emelie/tidy-html5
1
0
Fork 0
Code Issues Pull requests Releases Wiki Activity
tidy-html5/src/attrs.c

2404 lines
84 KiB
C
Raw Normal View History

new
2011-11-17 02:44:16 +00:00
/* attrs.c -- recognize HTML attributes
(c) 1998-2009 (W3C) MIT, ERCIM, Keio University
See tidy.h for the copyright notice.
*/
#include "tidy-int.h"
#include "attrs.h"
#include "message.h"
#include "tmbstr.h"
#include "utf8.h"
/*
Bind attribute types to procedures to check values.
You can add new procedures for better validation
and each procedure has access to the node in which
the attribute occurred as well as the attribute name
and its value.
By default, attributes are checked without regard
to the element they are found on. You have the choice
of making the procedure test which element is involved
or in writing methods for each element which controls
exactly how the attributes of that element are checked.
This latter approach is best for detecting the absence
of required attributes.
*/
static AttrCheck CheckAction;
static AttrCheck CheckScript;
static AttrCheck CheckName;
static AttrCheck CheckId;
- Add support for the `is` attribute. - Add support for autonomous custom elements.
2017-03-13 17:45:32 +00:00
static AttrCheck CheckIs;
new
2011-11-17 02:44:16 +00:00
static AttrCheck CheckAlign;
static AttrCheck CheckValign;
static AttrCheck CheckBool;
static AttrCheck CheckLength;
static AttrCheck CheckTarget;
static AttrCheck CheckFsubmit;
static AttrCheck CheckClear;
static AttrCheck CheckShape;
static AttrCheck CheckNumber;
static AttrCheck CheckScope;
static AttrCheck CheckColor;
static AttrCheck CheckVType;
static AttrCheck CheckScroll;
static AttrCheck CheckTextDir;
static AttrCheck CheckLang;
static AttrCheck CheckType;
Initial cut at RDFa support (again) New branch that implements support for RDFa attributes. Should be cleaner than my first attempt in PR #299 - also references issue #209
2015-11-16 17:29:23 +00:00
static AttrCheck CheckRDFaSafeCURIE;
static AttrCheck CheckRDFaTerm;
static AttrCheck CheckRDFaPrefix;
new
2011-11-17 02:44:16 +00:00
#define CH_PCDATA NULL
#define CH_CHARSET NULL
#define CH_TYPE CheckType
#define CH_XTYPE NULL
#define CH_CHARACTER NULL
#define CH_URLS NULL
#define CH_URL TY_(CheckUrl)
#define CH_SCRIPT CheckScript
#define CH_ALIGN CheckAlign
#define CH_VALIGN CheckValign
#define CH_COLOR CheckColor
#define CH_CLEAR CheckClear
#define CH_BORDER CheckBool /* kludge */
#define CH_LANG CheckLang
#define CH_BOOL CheckBool
#define CH_COLS NULL
#define CH_NUMBER CheckNumber
#define CH_LENGTH CheckLength
#define CH_COORDS NULL
#define CH_DATE NULL
#define CH_TEXTDIR CheckTextDir
#define CH_IDREFS NULL
#define CH_IDREF NULL
#define CH_IDDEF CheckId
- Add support for the `is` attribute. - Add support for autonomous custom elements.
2017-03-13 17:45:32 +00:00
#define CH_ISDEF CheckIs
new
2011-11-17 02:44:16 +00:00
#define CH_NAME CheckName
#define CH_TFRAME NULL
#define CH_FBORDER NULL
#define CH_MEDIA NULL
#define CH_FSUBMIT CheckFsubmit
#define CH_LINKTYPES NULL
#define CH_TRULES NULL
#define CH_SCOPE CheckScope
#define CH_SHAPE CheckShape
#define CH_SCROLL CheckScroll
#define CH_TARGET CheckTarget
#define CH_VTYPE CheckVType
#define CH_ACTION CheckAction
Initial cut at RDFa support (again) New branch that implements support for RDFa attributes. Should be cleaner than my first attempt in PR #299 - also references issue #209
2015-11-16 17:29:23 +00:00
#define CH_RDFAPREFIX CheckRDFaPrefix
#define CH_RDFASCURIE CheckRDFaSafeCURIE
#define CH_RDFASCURIES CheckRDFaSafeCURIE
#define CH_RDFATERM CheckRDFaTerm
#define CH_RDFATERMS CheckRDFaTerm
new
2011-11-17 02:44:16 +00:00
static const Attribute attribute_defs [] =
{
Because the previous effort #350 grew too fast and there was a LOT of side effects to my changes, I'm starting over with this. Comments in the PR thread. This commit reduces the size of attrdict.c while causing only a single errout regression that is justified.
2016-02-12 11:34:19 +00:00
{ TidyAttr_UNKNOWN, "unknown!", NULL },
{ TidyAttr_ABBR, "abbr", CH_PCDATA },
{ TidyAttr_ACCEPT, "accept", CH_XTYPE },
{ TidyAttr_ACCEPT_CHARSET, "accept-charset", CH_CHARSET },
{ TidyAttr_ACCESSKEY, "accesskey", CH_CHARACTER },
{ TidyAttr_ACTION, "action", CH_ACTION },
{ TidyAttr_ADD_DATE, "add_date", CH_PCDATA }, /* A */
{ TidyAttr_ALIGN, "align", CH_ALIGN }, /* varies by element */
{ TidyAttr_ALINK, "alink", CH_COLOR },
{ TidyAttr_ALLOWFULLSCREEN, "allowfullscreen", CH_BOOL },
{ TidyAttr_ALT, "alt", CH_PCDATA }, /* nowrap */
{ TidyAttr_ARCHIVE, "archive", CH_URLS }, /* space or comma separated list */
{ TidyAttr_AXIS, "axis", CH_PCDATA },
{ TidyAttr_BACKGROUND, "background", CH_URL },
{ TidyAttr_BGCOLOR, "bgcolor", CH_COLOR },
{ TidyAttr_BGPROPERTIES, "bgproperties", CH_PCDATA }, /* BODY "fixed" fixes background */
{ TidyAttr_BORDER, "border", CH_BORDER }, /* like LENGTH + "border" */
{ TidyAttr_BORDERCOLOR, "bordercolor", CH_COLOR }, /* used on TABLE */
{ TidyAttr_BOTTOMMARGIN, "bottommargin", CH_NUMBER }, /* used on BODY */
{ TidyAttr_CELLPADDING, "cellpadding", CH_LENGTH }, /* % or pixel values */
{ TidyAttr_CELLSPACING, "cellspacing", CH_LENGTH },
{ TidyAttr_CHAR, "char", CH_CHARACTER },
{ TidyAttr_CHAROFF, "charoff", CH_LENGTH },
{ TidyAttr_CHARSET, "charset", CH_CHARSET },
{ TidyAttr_CHECKED, "checked", CH_BOOL }, /* i.e. "checked" or absent */
{ TidyAttr_CITE, "cite", CH_URL },
{ TidyAttr_CLASS, "class", CH_PCDATA },
{ TidyAttr_CLASSID, "classid", CH_URL },
{ TidyAttr_CLEAR, "clear", CH_CLEAR }, /* BR: left, right, all */
{ TidyAttr_CODE, "code", CH_PCDATA }, /* APPLET */
{ TidyAttr_CODEBASE, "codebase", CH_URL }, /* OBJECT */
{ TidyAttr_CODETYPE, "codetype", CH_XTYPE }, /* OBJECT */
{ TidyAttr_COLOR, "color", CH_COLOR }, /* BASEFONT, FONT */
{ TidyAttr_COLS, "cols", CH_COLS }, /* TABLE & FRAMESET */
{ TidyAttr_COLSPAN, "colspan", CH_NUMBER },
{ TidyAttr_COMPACT, "compact", CH_BOOL }, /* lists */
{ TidyAttr_CONTENT, "content", CH_PCDATA },
{ TidyAttr_COORDS, "coords", CH_COORDS }, /* AREA, A */
{ TidyAttr_DATA, "data", CH_URL }, /* OBJECT */
{ TidyAttr_DATAFLD, "datafld", CH_PCDATA }, /* used on DIV, IMG */
{ TidyAttr_DATAFORMATAS, "dataformatas", CH_PCDATA }, /* used on DIV, IMG */
{ TidyAttr_DATAPAGESIZE, "datapagesize", CH_NUMBER }, /* used on DIV, IMG */
{ TidyAttr_DATASRC, "datasrc", CH_URL }, /* used on TABLE */
{ TidyAttr_DATETIME, "datetime", CH_DATE }, /* INS, DEL */
{ TidyAttr_DECLARE, "declare", CH_BOOL }, /* OBJECT */
{ TidyAttr_DEFER, "defer", CH_BOOL }, /* SCRIPT */
Make global attribute `dir` accept auto as well.
2018-04-04 10:29:30 +00:00
{ TidyAttr_DIR, "dir", CH_TEXTDIR }, /* ltr, rtl or auto */
Because the previous effort #350 grew too fast and there was a LOT of side effects to my changes, I'm starting over with this. Comments in the PR thread. This commit reduces the size of attrdict.c while causing only a single errout regression that is justified.
2016-02-12 11:34:19 +00:00
{ TidyAttr_DISABLED, "disabled", CH_BOOL }, /* form fields */
Address #598 - Add the download attribute to the anchor tag.
2017-09-29 19:30:43 +00:00
{ TidyAttr_DOWNLOAD, "download", CH_PCDATA }, /* anchor */
Because the previous effort #350 grew too fast and there was a LOT of side effects to my changes, I'm starting over with this. Comments in the PR thread. This commit reduces the size of attrdict.c while causing only a single errout regression that is justified.
2016-02-12 11:34:19 +00:00
{ TidyAttr_ENCODING, "encoding", CH_PCDATA }, /* <?xml?> */
{ TidyAttr_ENCTYPE, "enctype", CH_XTYPE }, /* FORM */
{ TidyAttr_FACE, "face", CH_PCDATA }, /* BASEFONT, FONT */
{ TidyAttr_FOR, "for", CH_IDREF }, /* LABEL */
{ TidyAttr_FRAME, "frame", CH_TFRAME }, /* TABLE */
{ TidyAttr_FRAMEBORDER, "frameborder", CH_FBORDER }, /* 0 or 1 */
{ TidyAttr_FRAMESPACING, "framespacing", CH_NUMBER },
{ TidyAttr_GRIDX, "gridx", CH_NUMBER }, /* TABLE Adobe golive*/
{ TidyAttr_GRIDY, "gridy", CH_NUMBER }, /* TABLE Adobe golive */
{ TidyAttr_HEADERS, "headers", CH_IDREFS }, /* table cells */
{ TidyAttr_HEIGHT, "height", CH_LENGTH }, /* pixels only for TH/TD */
{ TidyAttr_HREF, "href", CH_URL }, /* A, AREA, LINK and BASE */
{ TidyAttr_HREFLANG, "hreflang", CH_LANG }, /* A, LINK */
{ TidyAttr_HSPACE, "hspace", CH_NUMBER }, /* APPLET, IMG, OBJECT */
{ TidyAttr_HTTP_EQUIV, "http-equiv", CH_PCDATA }, /* META */
{ TidyAttr_ID, "id", CH_IDDEF },
- Add support for the `is` attribute. - Add support for autonomous custom elements.
2017-03-13 17:45:32 +00:00
{ TidyAttr_IS, "is", CH_ISDEF },
Because the previous effort #350 grew too fast and there was a LOT of side effects to my changes, I'm starting over with this. Comments in the PR thread. This commit reduces the size of attrdict.c while causing only a single errout regression that is justified.
2016-02-12 11:34:19 +00:00
{ TidyAttr_ISMAP, "ismap", CH_BOOL }, /* IMG */
{ TidyAttr_ITEMID, "itemid", CH_PCDATA },
{ TidyAttr_ITEMPROP, "itemprop", CH_PCDATA },
{ TidyAttr_ITEMREF, "itemref", CH_PCDATA },
{ TidyAttr_ITEMSCOPE, "itemscope", CH_BOOL },
{ TidyAttr_ITEMTYPE, "itemtype", CH_URL },
{ TidyAttr_LABEL, "label", CH_PCDATA }, /* OPT, OPTGROUP */
{ TidyAttr_LANG, "lang", CH_LANG },
{ TidyAttr_LANGUAGE, "language", CH_PCDATA }, /* SCRIPT */
{ TidyAttr_LAST_MODIFIED, "last_modified", CH_PCDATA }, /* A */
{ TidyAttr_LAST_VISIT, "last_visit", CH_PCDATA }, /* A */
{ TidyAttr_LEFTMARGIN, "leftmargin", CH_NUMBER }, /* used on BODY */
{ TidyAttr_LINK, "link", CH_COLOR }, /* BODY */
{ TidyAttr_LONGDESC, "longdesc", CH_URL }, /* IMG */
{ TidyAttr_LOWSRC, "lowsrc", CH_URL }, /* IMG */
{ TidyAttr_MARGINHEIGHT, "marginheight", CH_NUMBER }, /* FRAME, IFRAME, BODY */
{ TidyAttr_MARGINWIDTH, "marginwidth", CH_NUMBER }, /* ditto */
{ TidyAttr_MAXLENGTH, "maxlength", CH_NUMBER }, /* INPUT */
{ TidyAttr_MEDIA, "media", CH_MEDIA }, /* STYLE, LINK */
{ TidyAttr_METHOD, "method", CH_FSUBMIT }, /* FORM: get or post */
{ TidyAttr_MULTIPLE, "multiple", CH_BOOL }, /* SELECT */
{ TidyAttr_NAME, "name", CH_NAME },
{ TidyAttr_NOHREF, "nohref", CH_BOOL }, /* AREA */
{ TidyAttr_NORESIZE, "noresize", CH_BOOL }, /* FRAME */
{ TidyAttr_NOSHADE, "noshade", CH_BOOL }, /* HR */
{ TidyAttr_NOWRAP, "nowrap", CH_BOOL }, /* table cells */
{ TidyAttr_OBJECT, "object", CH_PCDATA }, /* APPLET */
{ TidyAttr_OnAFTERUPDATE, "onafterupdate", CH_SCRIPT },
{ TidyAttr_OnBEFOREUNLOAD, "onbeforeunload", CH_SCRIPT },
{ TidyAttr_OnBEFOREUPDATE, "onbeforeupdate", CH_SCRIPT },
{ TidyAttr_OnBLUR, "onblur", CH_SCRIPT }, /* event */
{ TidyAttr_OnCHANGE, "onchange", CH_SCRIPT }, /* event */
{ TidyAttr_OnCLICK, "onclick", CH_SCRIPT }, /* event */
{ TidyAttr_OnDATAAVAILABLE, "ondataavailable", CH_SCRIPT }, /* object, applet */
{ TidyAttr_OnDATASETCHANGED, "ondatasetchanged", CH_SCRIPT }, /* object, applet */
{ TidyAttr_OnDATASETCOMPLETE, "ondatasetcomplete", CH_SCRIPT },
{ TidyAttr_OnDBLCLICK, "ondblclick", CH_SCRIPT }, /* event */
{ TidyAttr_OnERRORUPDATE, "onerrorupdate", CH_SCRIPT }, /* form fields */
{ TidyAttr_OnFOCUS, "onfocus", CH_SCRIPT }, /* event */
{ TidyAttr_OnKEYDOWN, "onkeydown", CH_SCRIPT }, /* event */
{ TidyAttr_OnKEYPRESS, "onkeypress", CH_SCRIPT }, /* event */
{ TidyAttr_OnKEYUP, "onkeyup", CH_SCRIPT }, /* event */
{ TidyAttr_OnLOAD, "onload", CH_SCRIPT }, /* event */
{ TidyAttr_OnMOUSEDOWN, "onmousedown", CH_SCRIPT }, /* event */
{ TidyAttr_OnMOUSEMOVE, "onmousemove", CH_SCRIPT }, /* event */
{ TidyAttr_OnMOUSEOUT, "onmouseout", CH_SCRIPT }, /* event */
{ TidyAttr_OnMOUSEOVER, "onmouseover", CH_SCRIPT }, /* event */
{ TidyAttr_OnMOUSEUP, "onmouseup", CH_SCRIPT }, /* event */
{ TidyAttr_OnRESET, "onreset", CH_SCRIPT }, /* event */
{ TidyAttr_OnROWENTER, "onrowenter", CH_SCRIPT }, /* form fields */
{ TidyAttr_OnROWEXIT, "onrowexit", CH_SCRIPT }, /* form fields */
{ TidyAttr_OnSELECT, "onselect", CH_SCRIPT }, /* event */
{ TidyAttr_OnSUBMIT, "onsubmit", CH_SCRIPT }, /* event */
{ TidyAttr_OnUNLOAD, "onunload", CH_SCRIPT }, /* event */
{ TidyAttr_PROFILE, "profile", CH_URL }, /* HEAD */
{ TidyAttr_PROMPT, "prompt", CH_PCDATA }, /* ISINDEX */
{ TidyAttr_RBSPAN, "rbspan", CH_NUMBER }, /* ruby markup */
{ TidyAttr_READONLY, "readonly", CH_BOOL }, /* form fields */
{ TidyAttr_REL, "rel", CH_LINKTYPES },
{ TidyAttr_REV, "rev", CH_LINKTYPES },
{ TidyAttr_RIGHTMARGIN, "rightmargin", CH_NUMBER }, /* used on BODY */
{ TidyAttr_ROLE, "role", CH_PCDATA },
{ TidyAttr_ROWS, "rows", CH_NUMBER }, /* TEXTAREA */
{ TidyAttr_ROWSPAN, "rowspan", CH_NUMBER }, /* table cells */
{ TidyAttr_RULES, "rules", CH_TRULES }, /* TABLE */
{ TidyAttr_SCHEME, "scheme", CH_PCDATA }, /* META */
{ TidyAttr_SCOPE, "scope", CH_SCOPE }, /* table cells */
{ TidyAttr_SCROLLING, "scrolling", CH_SCROLL }, /* yes, no or auto */
{ TidyAttr_SELECTED, "selected", CH_BOOL }, /* OPTION */
{ TidyAttr_SHAPE, "shape", CH_SHAPE }, /* AREA, A */
{ TidyAttr_SHOWGRID, "showgrid", CH_BOOL }, /* TABLE Adobe golive */
{ TidyAttr_SHOWGRIDX, "showgridx", CH_BOOL }, /* TABLE Adobe golive*/
{ TidyAttr_SHOWGRIDY, "showgridy", CH_BOOL }, /* TABLE Adobe golive*/
{ TidyAttr_SIZE, "size", CH_NUMBER }, /* HR, FONT, BASEFONT, SELECT */
{ TidyAttr_SPAN, "span", CH_NUMBER }, /* COL, COLGROUP */
{ TidyAttr_SRC, "src", CH_URL }, /* IMG, FRAME, IFRAME */
{ TidyAttr_SRCSET, "srcset", CH_PCDATA }, /* IMG (HTML5) */
{ TidyAttr_STANDBY, "standby", CH_PCDATA }, /* OBJECT */
{ TidyAttr_START, "start", CH_NUMBER }, /* OL */
{ TidyAttr_STYLE, "style", CH_PCDATA },
{ TidyAttr_SUMMARY, "summary", CH_PCDATA }, /* TABLE */
{ TidyAttr_TABINDEX, "tabindex", CH_NUMBER }, /* fields, OBJECT and A */
{ TidyAttr_TARGET, "target", CH_TARGET }, /* names a frame/window */
{ TidyAttr_TEXT, "text", CH_COLOR }, /* BODY */
{ TidyAttr_TITLE, "title", CH_PCDATA }, /* text tool tip */
{ TidyAttr_TOPMARGIN, "topmargin", CH_NUMBER }, /* used on BODY */
{ TidyAttr_TRANSLATE, "translate", CH_BOOL }, /* HTML5 global attribute */
{ TidyAttr_TYPE, "type", CH_TYPE }, /* also used by SPACER */
{ TidyAttr_USEMAP, "usemap", CH_URL }, /* things with images */
{ TidyAttr_VALIGN, "valign", CH_VALIGN },
{ TidyAttr_VALUE, "value", CH_PCDATA },
{ TidyAttr_VALUETYPE, "valuetype", CH_VTYPE }, /* PARAM: data, ref, object */
{ TidyAttr_VERSION, "version", CH_PCDATA }, /* HTML <?xml?> */
{ TidyAttr_VLINK, "vlink", CH_COLOR }, /* BODY */
{ TidyAttr_VSPACE, "vspace", CH_NUMBER }, /* IMG, OBJECT, APPLET */
{ TidyAttr_WIDTH, "width", CH_LENGTH }, /* pixels only for TD/TH */
{ TidyAttr_WRAP, "wrap", CH_PCDATA }, /* textarea */
{ TidyAttr_XML_LANG, "xml:lang", CH_LANG }, /* XML language */
{ TidyAttr_XML_SPACE, "xml:space", CH_PCDATA }, /* XML white space */
new
2011-11-17 02:44:16 +00:00
/* todo: VERS_ALL is wrong! */
Because the previous effort #350 grew too fast and there was a LOT of side effects to my changes, I'm starting over with this. Comments in the PR thread. This commit reduces the size of attrdict.c while causing only a single errout regression that is justified.
2016-02-12 11:34:19 +00:00
{ TidyAttr_XMLNS, "xmlns", CH_PCDATA }, /* name space */
{ TidyAttr_EVENT, "event", CH_PCDATA }, /* reserved for <script> */
{ TidyAttr_METHODS, "methods", CH_PCDATA }, /* for <a>, never implemented */
{ TidyAttr_N, "n", CH_PCDATA }, /* for <nextid> */
{ TidyAttr_SDAFORM, "sdaform", CH_PCDATA }, /* SDATA attribute in HTML 2.0 */
{ TidyAttr_SDAPREF, "sdapref", CH_PCDATA }, /* SDATA attribute in HTML 2.0 */
{ TidyAttr_SDASUFF, "sdasuff", CH_PCDATA }, /* SDATA attribute in HTML 2.0 */
{ TidyAttr_URN, "urn", CH_PCDATA }, /* for <a>, never implemented */
new
2011-11-17 02:44:16 +00:00
Merge in TidyAttr_XML_LANG change.
2012-03-16 11:55:59 +00:00
/* HTML5 */
Because the previous effort #350 grew too fast and there was a LOT of side effects to my changes, I'm starting over with this. Comments in the PR thread. This commit reduces the size of attrdict.c while causing only a single errout regression that is justified.
2016-02-12 11:34:19 +00:00
{ TidyAttr_ASYNC, "async", CH_BOOL }, /* <script src="..." async> */
{ TidyAttr_AUTOCOMPLETE, "autocomplete", CH_PCDATA },
{ TidyAttr_AUTOFOCUS, "autofocus", CH_PCDATA },
{ TidyAttr_AUTOPLAY, "autoplay", CH_PCDATA },
{ TidyAttr_CHALLENGE, "challenge", CH_PCDATA },
{ TidyAttr_CONTENTEDITABLE, "contenteditable", CH_PCDATA },
{ TidyAttr_CONTEXTMENU, "contextmenu", CH_PCDATA },
{ TidyAttr_CONTROLS, "controls", CH_PCDATA },
Add support for the HTML5 "crossorigin" attribute This attribute can only be used on "link" elements. https://developer.mozilla.org/en-US/docs/Web/HTML/Element/link#Attributes
2016-06-08 02:20:10 +00:00
{ TidyAttr_CROSSORIGIN, "crossorigin", CH_PCDATA },
Because the previous effort #350 grew too fast and there was a LOT of side effects to my changes, I'm starting over with this. Comments in the PR thread. This commit reduces the size of attrdict.c while causing only a single errout regression that is justified.
2016-02-12 11:34:19 +00:00
{ TidyAttr_DEFAULT, "default", CH_PCDATA },
{ TidyAttr_DIRNAME, "dirname", CH_PCDATA },
{ TidyAttr_DRAGGABLE, "draggable", CH_PCDATA },
{ TidyAttr_DROPZONE, "dropzone", CH_PCDATA },
{ TidyAttr_FORM, "form", CH_PCDATA },
{ TidyAttr_FORMACTION, "formaction", CH_PCDATA },
{ TidyAttr_FORMENCTYPE, "formenctype", CH_PCDATA },
{ TidyAttr_FORMMETHOD, "formmethod", CH_PCDATA },
{ TidyAttr_FORMNOVALIDATE, "formnovalidate", CH_PCDATA },
{ TidyAttr_FORMTARGET, "formtarget", CH_PCDATA },
{ TidyAttr_HIDDEN, "hidden", CH_PCDATA },
{ TidyAttr_HIGH, "high", CH_PCDATA },
{ TidyAttr_ICON, "icon", CH_PCDATA },
{ TidyAttr_KEYTYPE, "keytype", CH_PCDATA },
{ TidyAttr_KIND, "kind", CH_PCDATA },
{ TidyAttr_LIST, "list", CH_PCDATA },
{ TidyAttr_LOOP, "loop", CH_PCDATA },
{ TidyAttr_LOW, "low", CH_PCDATA },
{ TidyAttr_MANIFEST, "manifest", CH_PCDATA },
{ TidyAttr_MAX, "max", CH_PCDATA },
{ TidyAttr_MEDIAGROUP, "mediagroup", CH_PCDATA },
{ TidyAttr_MIN, "min", CH_PCDATA },
{ TidyAttr_NOVALIDATE, "novalidate", CH_PCDATA },
{ TidyAttr_OPEN, "open", CH_PCDATA },
{ TidyAttr_OPTIMUM, "optimum", CH_PCDATA },
{ TidyAttr_OnABORT, "onabort", CH_PCDATA },
{ TidyAttr_OnAFTERPRINT, "onafterprint", CH_PCDATA },
{ TidyAttr_OnBEFOREPRINT, "onbeforeprint", CH_PCDATA },
{ TidyAttr_OnCANPLAY, "oncanplay", CH_PCDATA },
{ TidyAttr_OnCANPLAYTHROUGH, "oncanplaythrough", CH_PCDATA },
{ TidyAttr_OnCONTEXTMENU, "oncontextmenu", CH_PCDATA },
{ TidyAttr_OnCUECHANGE, "oncuechange", CH_PCDATA },
{ TidyAttr_OnDRAG, "ondrag", CH_PCDATA },
{ TidyAttr_OnDRAGEND, "ondragend", CH_PCDATA },
{ TidyAttr_OnDRAGENTER, "ondragenter", CH_PCDATA },
{ TidyAttr_OnDRAGLEAVE, "ondragleave", CH_PCDATA },
{ TidyAttr_OnDRAGOVER, "ondragover", CH_PCDATA },
{ TidyAttr_OnDRAGSTART, "ondragstart", CH_PCDATA },
{ TidyAttr_OnDROP, "ondrop", CH_PCDATA },
{ TidyAttr_OnDURATIONCHANGE, "ondurationchange", CH_PCDATA },
{ TidyAttr_OnEMPTIED, "onemptied", CH_PCDATA },
{ TidyAttr_OnENDED, "onended", CH_PCDATA },
{ TidyAttr_OnERROR, "onerror", CH_PCDATA },
{ TidyAttr_OnHASHCHANGE, "onhashchange", CH_PCDATA },
{ TidyAttr_OnINPUT, "oninput", CH_PCDATA },
{ TidyAttr_OnINVALID, "oninvalid", CH_PCDATA },
{ TidyAttr_OnLOADEDDATA, "onloadeddata", CH_PCDATA },
{ TidyAttr_OnLOADEDMETADATA, "onloadedmetadata", CH_PCDATA },
{ TidyAttr_OnLOADSTART, "onloadstart", CH_PCDATA },
{ TidyAttr_OnMESSAGE, "onmessage", CH_PCDATA },
{ TidyAttr_OnMOUSEWHEEL, "onmousewheel", CH_PCDATA },
{ TidyAttr_OnOFFLINE, "onoffline", CH_PCDATA },
{ TidyAttr_OnONLINE, "ononline", CH_PCDATA },
{ TidyAttr_OnPAGEHIDE, "onpagehide", CH_PCDATA },
{ TidyAttr_OnPAGESHOW, "onpageshow", CH_PCDATA },
{ TidyAttr_OnPAUSE, "onpause", CH_PCDATA },
{ TidyAttr_OnPLAY, "onplay", CH_PCDATA },
{ TidyAttr_OnPLAYING, "onplaying", CH_PCDATA },
{ TidyAttr_OnPOPSTATE, "onpopstate", CH_PCDATA },
{ TidyAttr_OnPROGRESS, "onprogress", CH_PCDATA },
{ TidyAttr_OnRATECHANGE, "onratechange", CH_PCDATA },
{ TidyAttr_OnREADYSTATECHANGE, "onreadystatechange", CH_PCDATA },
{ TidyAttr_OnREDO, "onredo", CH_PCDATA },
{ TidyAttr_OnRESIZE, "onresize", CH_PCDATA },
{ TidyAttr_OnSCROLL, "onscroll", CH_PCDATA },
{ TidyAttr_OnSEEKED, "onseeked", CH_PCDATA },
{ TidyAttr_OnSEEKING, "onseeking", CH_PCDATA },
{ TidyAttr_OnSHOW, "onshow", CH_PCDATA },
{ TidyAttr_OnSTALLED, "onstalled", CH_PCDATA },
{ TidyAttr_OnSTORAGE, "onstorage", CH_PCDATA },
{ TidyAttr_OnSUSPEND, "onsuspend", CH_PCDATA },
{ TidyAttr_OnTIMEUPDATE, "ontimeupdate", CH_PCDATA },
{ TidyAttr_OnUNDO, "onundo", CH_PCDATA },
{ TidyAttr_OnVOLUMECHANGE, "onvolumechange", CH_PCDATA },
{ TidyAttr_OnWAITING, "onwaiting", CH_PCDATA },
{ TidyAttr_PATTERN, "pattern", CH_PCDATA },
{ TidyAttr_PLACEHOLDER, "placeholder", CH_PCDATA },
{ TidyAttr_POSTER, "poster", CH_PCDATA },
{ TidyAttr_PRELOAD, "preload", CH_PCDATA },
{ TidyAttr_PUBDATE, "pubdate", CH_PCDATA },
{ TidyAttr_RADIOGROUP, "radiogroup", CH_PCDATA },
{ TidyAttr_REQUIRED, "required", CH_PCDATA },
{ TidyAttr_REVERSED, "reversed", CH_PCDATA },
{ TidyAttr_SANDBOX, "sandbox", CH_PCDATA },
{ TidyAttr_SCOPED, "scoped", CH_PCDATA },
{ TidyAttr_SEAMLESS, "seamless", CH_PCDATA },
{ TidyAttr_SIZES, "sizes", CH_PCDATA },
{ TidyAttr_SPELLCHECK, "spellcheck", CH_PCDATA },
{ TidyAttr_SRCDOC, "srcdoc", CH_PCDATA },
{ TidyAttr_SRCLANG, "srclang", CH_PCDATA },
{ TidyAttr_STEP, "step", CH_PCDATA },
new
2011-11-17 02:44:16 +00:00
Add support for aria attributes Reference: http://dev.w3.org/html5/markup/aria/aria.html#aria-attrs-all
2013-06-28 23:20:02 +00:00
/* HTML5 Aria Attributes */
{ TidyAttr_ARIA_ACTIVEDESCENDANT, "aria-activedescendant", CH_PCDATA },
{ TidyAttr_ARIA_ATOMIC, "aria-atomic", CH_PCDATA },
{ TidyAttr_ARIA_AUTOCOMPLETE, "aria-autocomplete", CH_PCDATA },
{ TidyAttr_ARIA_BUSY, "aria-busy", CH_PCDATA },
{ TidyAttr_ARIA_CHECKED, "aria-checked", CH_PCDATA },
{ TidyAttr_ARIA_CONTROLS, "aria-controls", CH_PCDATA },
{ TidyAttr_ARIA_DESCRIBEDBY, "aria-describedby", CH_PCDATA },
{ TidyAttr_ARIA_DISABLED, "aria-disabled", CH_PCDATA },
{ TidyAttr_ARIA_DROPEFFECT, "aria-dropeffect", CH_PCDATA },
{ TidyAttr_ARIA_EXPANDED, "aria-expanded", CH_PCDATA },
{ TidyAttr_ARIA_FLOWTO, "aria-flowto", CH_PCDATA },
{ TidyAttr_ARIA_GRABBED, "aria-grabbed", CH_PCDATA },
{ TidyAttr_ARIA_HASPOPUP, "aria-haspopup", CH_PCDATA },
{ TidyAttr_ARIA_HIDDEN, "aria-hidden", CH_PCDATA },
{ TidyAttr_ARIA_INVALID, "aria-invalid", CH_PCDATA },
{ TidyAttr_ARIA_LABEL, "aria-label", CH_PCDATA },
{ TidyAttr_ARIA_LABELLEDBY, "aria-labelledby", CH_PCDATA },
{ TidyAttr_ARIA_LEVEL, "aria-level", CH_PCDATA },
{ TidyAttr_ARIA_LIVE, "aria-live", CH_PCDATA },
{ TidyAttr_ARIA_MULTILINE, "aria-multiline", CH_PCDATA },
{ TidyAttr_ARIA_MULTISELECTABLE, "aria-multiselectable", CH_PCDATA },
Updated Aria attributes to geoffmcl's added tags; added missing aria-orientation.
2014-11-22 09:39:17 +00:00
{ TidyAttr_ARIA_ORIENTATION, "aria-orientation", CH_PCDATA },
Add support for aria attributes Reference: http://dev.w3.org/html5/markup/aria/aria.html#aria-attrs-all
2013-06-28 23:20:02 +00:00
{ TidyAttr_ARIA_OWNS, "aria-owns", CH_PCDATA },
{ TidyAttr_ARIA_POSINSET, "aria-posinset", CH_PCDATA },
{ TidyAttr_ARIA_PRESSED, "aria-pressed", CH_PCDATA },
{ TidyAttr_ARIA_READONLY, "aria-readonly", CH_PCDATA },
{ TidyAttr_ARIA_RELEVANT, "aria-relevant", CH_PCDATA },
{ TidyAttr_ARIA_REQUIRED, "aria-required", CH_PCDATA },
{ TidyAttr_ARIA_SELECTED, "aria-selected", CH_PCDATA },
{ TidyAttr_ARIA_SETSIZE, "aria-setsize", CH_PCDATA },
{ TidyAttr_ARIA_SORT, "aria-sort", CH_PCDATA },
{ TidyAttr_ARIA_VALUEMAX, "aria-valuemax", CH_PCDATA },
{ TidyAttr_ARIA_VALUEMIN, "aria-valuemin", CH_PCDATA },
{ TidyAttr_ARIA_VALUENOW, "aria-valuenow", CH_PCDATA },
{ TidyAttr_ARIA_VALUETEXT, "aria-valuetext", CH_PCDATA },
Because the previous effort #350 grew too fast and there was a LOT of side effects to my changes, I'm starting over with this. Comments in the PR thread. This commit reduces the size of attrdict.c while causing only a single errout regression that is justified.
2016-02-12 11:34:19 +00:00
{ TidyAttr_X, "x", CH_PCDATA }, /* for <svg> */
{ TidyAttr_Y, "y", CH_PCDATA }, /* for <svg> */
Issue #70 - Initial implmentation of SVG support. An immense thanks to Ger Hobbelt who had already done this in his github.com/GerHobbelt/htmltidy fork. The two sources have diverges so was not a simple cut an paste. But again thanks Ger for this.
2015-02-02 16:25:49 +00:00
{ TidyAttr_VIEWBOX, "viewbox", CH_PCDATA }, /* for <svg> */
{ TidyAttr_PRESERVEASPECTRATIO, "preserveaspectratio", CH_PCDATA }, /* for <svg> */
{ TidyAttr_ZOOMANDPAN, "zoomandpan", CH_PCDATA }, /* for <svg> */
{ TidyAttr_BASEPROFILE, "baseprofile", CH_PCDATA }, /* for <svg> */
{ TidyAttr_CONTENTSCRIPTTYPE, "contentscripttype", CH_PCDATA }, /* for <svg> */
{ TidyAttr_CONTENTSTYLETYPE, "contentstyletype", CH_PCDATA }, /* for <svg> */
Because the previous effort #350 grew too fast and there was a LOT of side effects to my changes, I'm starting over with this. Comments in the PR thread. This commit reduces the size of attrdict.c while causing only a single errout regression that is justified.
2016-02-12 11:34:19 +00:00
{ TidyAttr_DISPLAY, "display", CH_PCDATA }, /* on MATH tag (html5) */
Issue #70 - Initial implmentation of SVG support. An immense thanks to Ger Hobbelt who had already done this in his github.com/GerHobbelt/htmltidy fork. The two sources have diverges so was not a simple cut an paste. But again thanks Ger for this.
2015-02-02 16:25:49 +00:00
Initial cut at RDFa support (again) New branch that implements support for RDFa attributes. Should be cleaner than my first attempt in PR #299 - also references issue #209
2015-11-16 17:29:23 +00:00
/* RDFa Attributes */
Because the previous effort #350 grew too fast and there was a LOT of side effects to my changes, I'm starting over with this. Comments in the PR thread. This commit reduces the size of attrdict.c while causing only a single errout regression that is justified.
2016-02-12 11:34:19 +00:00
{ TidyAttr_ABOUT, "about", CH_RDFASCURIE },
{ TidyAttr_DATATYPE, "datatype", CH_RDFATERM },
{ TidyAttr_INLIST, "inlist", CH_BOOL },
{ TidyAttr_PREFIX, "prefix", CH_RDFAPREFIX },
{ TidyAttr_PROPERTY, "property", CH_RDFATERMS },
{ TidyAttr_RESOURCE, "resource", CH_RDFASCURIE },
{ TidyAttr_TYPEOF, "typeof", CH_RDFATERMS },
{ TidyAttr_VOCAB, "vocab", CH_URL },
Initial cut at RDFa support (again) New branch that implements support for RDFa attributes. Should be cleaner than my first attempt in PR #299 - also references issue #209
2015-11-16 17:29:23 +00:00
Add support for the "integrity" attribute This attribute may be used on "link" and "script" elements. See http://www.w3.org/TR/2016/REC-SRI-20160623/#element-interface-extensions
2016-07-24 14:24:30 +00:00
{ TidyAttr_INTEGRITY, "integrity", CH_PCDATA },
Add support for link 'as' attribute (closes #449)
2016-08-23 08:36:25 +00:00
/* Preload spec: https://www.w3.org/TR/preload/ */
{ TidyAttr_AS, "as", CH_PCDATA },
Add optional xmlns:xlink attributes as valid to support inline svg fixes #478
2017-01-08 23:38:16 +00:00
/* for xmlns:xlink in <svg> */
{ TidyAttr_XMLNSXLINK, "xmlns:xlink", CH_URL },
new
2011-11-17 02:44:16 +00:00
/* this must be the final entry */
Because the previous effort #350 grew too fast and there was a LOT of side effects to my changes, I'm starting over with this. Comments in the PR thread. This commit reduces the size of attrdict.c while causing only a single errout regression that is justified.
2016-02-12 11:34:19 +00:00
{ N_TIDY_ATTRIBS, NULL, NULL }
new
2011-11-17 02:44:16 +00:00
};
static uint AttributeVersions(Node* node, AttVal* attval)
{
uint i;
Hard-coded checks are redundant with updates to `attrdict.c`.
2016-02-12 12:44:03 +00:00
/* Override or add to items in attrdict.c */
add allowfullscreen attribute exception
2015-01-18 19:59:27 +00:00
if (attval && attval->attribute) {
Hard-coded checks are redundant with updates to `attrdict.c`.
2016-02-12 12:44:03 +00:00
/* HTML5 data-* attributes can't be added generically; handle here. */
new
2011-11-17 02:44:16 +00:00
if (TY_(tmbstrncmp)(attval->attribute, "data-", 5) == 0)
return (XH50 | HT50);
add allowfullscreen attribute exception
2015-01-18 19:59:27 +00:00
}
new
2011-11-17 02:44:16 +00:00
/* TODO: maybe this should return VERS_PROPRIETARY instead? */
if (!attval || !attval->dict)
return VERS_UNKNOWN;
if (!(!node || !node->tag || !node->tag->attrvers))
for (i = 0; node->tag->attrvers[i].attribute; ++i)
if (node->tag->attrvers[i].attribute == attval->dict->id)
return node->tag->attrvers[i].versions;
return VERS_PROPRIETARY;
}
/* return the version of the attribute "id" of element "node" */
uint TY_(NodeAttributeVersions)( Node* node, TidyAttrId id )
{
uint i;
if (!node || !node->tag || !node->tag->attrvers)
return VERS_UNKNOWN;
for (i = 0; node->tag->attrvers[i].attribute; ++i)
if (node->tag->attrvers[i].attribute == id)
return node->tag->attrvers[i].versions;
return VERS_UNKNOWN;
}
Phase 2 This is a MUCH SANER approach to what I was trying to do (now that I screwed up enough internals to understand some of them! At this point there are zero exit state reversions, and zero markup reversions! There are still 21 errout reversions; I'll annotate and adjust as necessary.
2016-02-13 03:31:16 +00:00
/* returns true if the element is a W3C defined element
* but the element/attribute combination is not. We're
* only defining as "proprietary" items that are not in
* the element's AttrVersion structure.
*/
Bool TY_(AttributeIsProprietary)(Node* node, AttVal* attval)
new
2011-11-17 02:44:16 +00:00
{
if (!node || !attval)
return no;
if (!node->tag)
return no;
if (!(node->tag->versions & VERS_ALL))
return no;
if (AttributeVersions(node, attval) & VERS_ALL)
return no;
return yes;
}
Phase 2 This is a MUCH SANER approach to what I was trying to do (now that I screwed up enough internals to understand some of them! At this point there are zero exit state reversions, and zero markup reversions! There are still 21 errout reversions; I'll annotate and adjust as necessary.
2016-02-13 03:31:16 +00:00
/* returns true if the element is a W3C defined element
* but the element/attribute combination is not. We're
* considering it a mismatch if the document version
* does not allow the attribute as called out in its
* AttrVersion structure.
*/
Bool TY_(AttributeIsMismatched)(Node* node, AttVal* attval, TidyDocImpl* doc)
{
uint doctype;
if (!node || !attval)
return no;
if (!node->tag)
return no;
if (!(node->tag->versions & VERS_ALL))
return no;
doctype = doc->lexer->versionEmitted == 0 ? doc->lexer->doctype : doc->lexer->versionEmitted;
if (AttributeVersions(node, attval) & doctype)
return no;
return yes;
}
new
2011-11-17 02:44:16 +00:00
/* used by CheckColor() */
struct _colors
{
ctmbstr name;
ctmbstr hex;
};
static const struct _colors colors[] =
{
{ "black", "#000000" },
{ "green", "#008000" },
{ "silver", "#C0C0C0" },
{ "lime", "#00FF00" },
{ "gray", "#808080" },
{ "olive", "#808000" },
{ "white", "#FFFFFF" },
{ "yellow", "#FFFF00" },
{ "maroon", "#800000" },
{ "navy", "#000080" },
{ "red", "#FF0000" },
{ "blue", "#0000FF" },
{ "purple", "#800080" },
{ "teal", "#008080" },
{ "fuchsia", "#FF00FF" },
{ "aqua", "#00FFFF" },
{ NULL, NULL }
};
static ctmbstr GetColorCode(ctmbstr name)
{
uint i;
for (i = 0; colors[i].name; ++i)
if (TY_(tmbstrcasecmp)(name, colors[i].name) == 0)
return colors[i].hex;
return NULL;
}
static ctmbstr GetColorName(ctmbstr code)
{
uint i;
for (i = 0; colors[i].name; ++i)
if (TY_(tmbstrcasecmp)(code, colors[i].hex) == 0)
return colors[i].name;
return NULL;
}
static uint attrsHash(ctmbstr s)
{
uint hashval;
for (hashval = 0; *s != '\0'; s++)
hashval = *s + 31*hashval;
return hashval % ATTRIBUTE_HASH_SIZE;
}
static const Attribute *attrsInstall(TidyDocImpl* doc, TidyAttribImpl * attribs,
const Attribute* old)
{
AttrHash *np;
uint hashval;
if (old)
{
np = (AttrHash *)TidyDocAlloc(doc, sizeof(*np));
np->attr = old;
hashval = attrsHash(old->name);
np->next = attribs->hashtab[hashval];
attribs->hashtab[hashval] = np;
}
return old;
}
static void attrsRemoveFromHash( TidyDocImpl* doc, TidyAttribImpl *attribs,
ctmbstr s )
{
uint h = attrsHash(s);
AttrHash *p, *prev = NULL;
for (p = attribs->hashtab[h]; p && p->attr; p = p->next)
{
if (TY_(tmbstrcmp)(s, p->attr->name) == 0)
{
AttrHash* next = p->next;
if ( prev )
prev->next = next;
else
attribs->hashtab[h] = next;
TidyDocFree(doc, p);
return;
}
prev = p;
}
}
static void attrsEmptyHash( TidyDocImpl* doc, TidyAttribImpl * attribs )
{
AttrHash *dict, *next;
uint i;
for (i = 0; i < ATTRIBUTE_HASH_SIZE; ++i)
{
dict = attribs->hashtab[i];
while(dict)
{
next = dict->next;
TidyDocFree(doc, dict);
dict = next;
}
attribs->hashtab[i] = NULL;
}
}
static const Attribute* attrsLookup(TidyDocImpl* doc,
TidyAttribImpl* ARG_UNUSED(attribs),
ctmbstr atnam)
{
const Attribute *np;
const AttrHash *p;
if (!atnam)
return NULL;
for (p = attribs->hashtab[attrsHash(atnam)]; p && p->attr; p = p->next)
Case insensitive compare is safe here, and prevents erroneous propriertary attribute errors.
2017-05-12 12:28:11 +00:00
if (TY_(tmbstrcasecmp)(atnam, p->attr->name) == 0)
new
2011-11-17 02:44:16 +00:00
return p->attr;
for (np = attribute_defs; np && np->name; ++np)
Case insensitive compare is safe here, and prevents erroneous propriertary attribute errors.
2017-05-12 12:28:11 +00:00
if (TY_(tmbstrcasecmp)(atnam, np->name) == 0)
new
2011-11-17 02:44:16 +00:00
return attrsInstall(doc, attribs, np);
return NULL;
}
/* Locate attributes by type */
AttVal* TY_(AttrGetById)( Node* node, TidyAttrId id )
{
AttVal* av;
for ( av = node->attributes; av; av = av->next )
{
if ( AttrIsId(av, id) )
return av;
}
return NULL;
}
/* public method for finding attribute definition by name */
const Attribute* TY_(FindAttribute)( TidyDocImpl* doc, AttVal *attval )
{
if ( attval )
return attrsLookup( doc, &doc->attribs, attval->attribute );
return NULL;
}
AttVal* TY_(GetAttrByName)( Node *node, ctmbstr name )
{
AttVal *attr;
for (attr = node->attributes; attr != NULL; attr = attr->next)
{
if (attr->attribute && TY_(tmbstrcmp)(attr->attribute, name) == 0)
break;
}
return attr;
}
Let's actually commit the -gdoc feature this time.
2012-06-20 07:55:42 +00:00
void TY_(DropAttrByName)( TidyDocImpl* doc, Node *node, ctmbstr name )
{
AttVal *attr, *prev = NULL, *next;
for (attr = node->attributes; attr != NULL; prev = attr, attr = next)
{
next = attr->next;
if (attr->attribute && TY_(tmbstrcmp)(attr->attribute, name) == 0)
{
if (prev)
prev->next = next;
else
node->attributes = next;
TY_(FreeAttribute)( doc, attr );
break;
}
}
}
new
2011-11-17 02:44:16 +00:00
AttVal* TY_(AddAttribute)( TidyDocImpl* doc,
Node *node, ctmbstr name, ctmbstr value )
{
AttVal *av = TY_(NewAttribute)(doc);
av->delim = '"';
av->attribute = TY_(tmbstrdup)(doc->allocator, name);
if (value)
av->value = TY_(tmbstrdup)(doc->allocator, value);
else
av->value = NULL;
av->dict = attrsLookup(doc, &doc->attribs, name);
TY_(InsertAttributeAtEnd)(node, av);
return av;
}
AttVal* TY_(RepairAttrValue)(TidyDocImpl* doc, Node* node, ctmbstr name, ctmbstr value)
{
AttVal* old = TY_(GetAttrByName)(node, name);
if (old)
{
if (old->value)
TidyDocFree(doc, old->value);
if (value)
old->value = TY_(tmbstrdup)(doc->allocator, value);
else
old->value = NULL;
return old;
}
else
return TY_(AddAttribute)(doc, node, name, value);
}
Initial commit of #434; WIP
2017-09-23 02:14:59 +00:00
New config option now drives the list; memory deallocated.
2017-09-23 11:20:31 +00:00
void TY_(FreeAttrPriorityList)( TidyDocImpl* doc )
{
PriorityAttribs *priorities = &(doc->attribs.priorityAttribs);
if ( priorities->list )
{
uint i = 0;
while ( priorities->list[i] != NULL )
{
TidyFree( doc->allocator, priorities->list[i] );
i++;
}
TidyFree( doc->allocator, priorities->list );
}
}
Spacing consistency.
2017-10-08 14:25:09 +00:00
Initial commit of #434; WIP
2017-09-23 02:14:59 +00:00
void TY_(DefinePriorityAttribute)(TidyDocImpl* doc, ctmbstr name)
{
enum { capacity = 10 };
PriorityAttribs *priorities = &(doc->attribs.priorityAttribs);
if ( !priorities->list )
{
New config option now drives the list; memory deallocated.
2017-09-23 11:20:31 +00:00
priorities->list = TidyAlloc(doc->allocator, sizeof(ctmbstr) * capacity );
Initial commit of #434; WIP
2017-09-23 02:14:59 +00:00
priorities->list[0] = NULL;
priorities->capacity = capacity;
priorities->count = 0;
}
if ( priorities->count >= priorities->capacity )
{
priorities->capacity = priorities->capacity * 2;
priorities->list = realloc( priorities->list, sizeof(tmbstr) * priorities->capacity + 1 );
}
priorities->list[priorities->count] = TY_(tmbstrdup)( doc->allocator, name);
priorities->count++;
priorities->list[priorities->count] = NULL;
}
Add missing iterators for recently added options mute and priority attributes.
2017-11-19 15:21:46 +00:00
TidyIterator TY_(getPriorityAttrList)( TidyDocImpl* doc )
{
PriorityAttribs *priorities = &(doc->attribs.priorityAttribs);
size_t result = priorities->count > 0 ? 1 : 0;
return (TidyIterator) result;
}
ctmbstr TY_(getNextPriorityAttr)( TidyDocImpl* doc, TidyIterator* iter )
{
PriorityAttribs *priorities = &(doc->attribs.priorityAttribs);
size_t index;
ctmbstr result = NULL;
assert( iter != NULL );
index = (size_t)*iter;
Fixed logical issue with new iterators.
2017-11-19 16:02:13 +00:00
if ( index > 0 && index <= priorities->count )
Add missing iterators for recently added options mute and priority attributes.
2017-11-19 15:21:46 +00:00
{
result = priorities->list[index-1];
index++;
}
*iter = (TidyIterator) ( index <= priorities->count ? index : (size_t)0 );
return result;
}
new
2011-11-17 02:44:16 +00:00
static Bool CheckAttrType( TidyDocImpl* doc,
ctmbstr attrname, AttrCheck type )
{
const Attribute* np = attrsLookup( doc, &doc->attribs, attrname );
return (Bool)( np && np->attrchk == type );
}
Bool TY_(IsUrl)( TidyDocImpl* doc, ctmbstr attrname )
{
return CheckAttrType( doc, attrname, CH_URL );
}
/*
Bool IsBool( TidyDocImpl* doc, ctmbstr attrname )
{
return CheckAttrType( doc, attrname, CH_BOOL );
}
*/
Bool TY_(IsScript)( TidyDocImpl* doc, ctmbstr attrname )
{
return CheckAttrType( doc, attrname, CH_SCRIPT );
}
/* may id or name serve as anchor? */
Bool TY_(IsAnchorElement)( TidyDocImpl* ARG_UNUSED(doc), Node* node)
{
TidyTagId tid = TagId( node );
if ( tid == TidyTag_A ||
tid == TidyTag_APPLET ||
tid == TidyTag_FORM ||
tid == TidyTag_FRAME ||
tid == TidyTag_IFRAME ||
tid == TidyTag_IMG ||
tid == TidyTag_MAP )
return yes;
return no;
}
/*
In CSS1, selectors can contain only the characters A-Z, 0-9,
and Unicode characters 161-255, plus dash (-); they cannot start
with a dash or a digit; they can also contain escaped characters
and any Unicode character as a numeric code (see next item).
The backslash followed by at most four hexadecimal digits
(0..9A..F) stands for the Unicode character with that number.
Any character except a hexadecimal digit can be escaped to remove
its special meaning, by putting a backslash in front.
#508936 - CSS class naming for -clean option
*/
Bool TY_(IsCSS1Selector)( ctmbstr buf )
{
Bool valid = yes;
int esclen = 0;
byte c;
int pos;
for ( pos=0; valid && (c = *buf++); ++pos )
{
if ( c == '\\' )
{
esclen = 1; /* ab\555\444 is 4 chars {'a', 'b', \555, \444} */
}
else if ( isdigit( c ) )
{
/* Digit not 1st, unless escaped (Max length "\112F") */
if ( esclen > 0 )
valid = ( ++esclen < 6 );
if ( valid )
valid = ( pos>0 || esclen>0 );
}
else
{
valid = (
esclen > 0 /* Escaped? Anything goes. */
|| ( pos>0 && c == '-' ) /* Dash cannot be 1st char */
|| isalpha(c) /* a-z, A-Z anywhere */
|| ( c >= 161 ) /* Unicode 161-255 anywhere */
);
esclen = 0;
}
}
return valid;
}
/* free single anchor */
static void FreeAnchor(TidyDocImpl* doc, Anchor *a)
{
if ( a )
TidyDocFree( doc, a->name );
TidyDocFree( doc, a );
}
Applied hash table optimisation to RemoveAnchorByNode. This function now takes the anchor name as a parameter, so it can look in the correct bin. In the case of FreeAttrs, we have the name already (since we found a name or id attribute). In the case of FixAnchors, the anchor name could come from either the name or id attribute, so we call the function separately for each case, passing the appropriate attribute value.
2012-08-20 03:06:30 +00:00
static uint anchorNameHash(ctmbstr s)
{
Issue #149 - avoid crash on null attr value
2015-02-03 12:38:20 +00:00
uint hashval = 0;
/* Issue #149 - an inferred name can be null. avoid crash */
if (s)
{
for ( ; *s != '\0'; s++) {
tmbchar c = TY_(ToLower)( *s );
hashval = c + 31*hashval;
}
Applied hash table optimisation to RemoveAnchorByNode. This function now takes the anchor name as a parameter, so it can look in the correct bin. In the case of FreeAttrs, we have the name already (since we found a name or id attribute). In the case of FixAnchors, the anchor name could come from either the name or id attribute, so we call the function separately for each case, passing the appropriate attribute value.
2012-08-20 03:06:30 +00:00
}
return hashval % ANCHOR_HASH_SIZE;
}
Issue #185 - Treat elements ids as case-sensitive if in HTML5 mode
2015-03-13 18:47:28 +00:00
/*\
* New Service for HTML5
* Issue #185 - Treat elements ids as case-sensitive
* if in HTML5 modes, make hash of value AS IS!
\*/
static uint anchorNameHash5(ctmbstr s)
{
uint hashval = 0;
/* Issue #149 - an inferred name can be null. avoid crash */
if (s)
{
for ( ; *s != '\0'; s++) {
tmbchar c = *s;
hashval = c + 31*hashval;
}
}
return hashval % ANCHOR_HASH_SIZE;
}
/*\
* removes anchor for specific node
* Issue #185 - Treat elements ids as case-sensitive
* if in HTML5 modes, make hash of value AS IS!
\*/
Applied hash table optimisation to RemoveAnchorByNode. This function now takes the anchor name as a parameter, so it can look in the correct bin. In the case of FreeAttrs, we have the name already (since we found a name or id attribute). In the case of FixAnchors, the anchor name could come from either the name or id attribute, so we call the function separately for each case, passing the appropriate attribute value.
2012-08-20 03:06:30 +00:00
void TY_(RemoveAnchorByNode)( TidyDocImpl* doc, ctmbstr name, Node *node )
new
2011-11-17 02:44:16 +00:00
{
TidyAttribImpl* attribs = &doc->attribs;
Anchor *delme = NULL, *curr, *prev = NULL;
Issue #185 - Treat elements ids as case-sensitive if in HTML5 mode
2015-03-13 18:47:28 +00:00
uint h;
if (TY_(HTMLVersion)(doc) == HT50)
h = anchorNameHash5(name);
else
h = anchorNameHash(name);
new
2011-11-17 02:44:16 +00:00
Applied hash table optimisation to RemoveAnchorByNode. This function now takes the anchor name as a parameter, so it can look in the correct bin. In the case of FreeAttrs, we have the name already (since we found a name or id attribute). In the case of FixAnchors, the anchor name could come from either the name or id attribute, so we call the function separately for each case, passing the appropriate attribute value.
2012-08-20 03:06:30 +00:00
for ( curr=attribs->anchor_hash[h]; curr!=NULL; curr=curr->next )
new
2011-11-17 02:44:16 +00:00
{
if ( curr->node == node )
{
if ( prev )
prev->next = curr->next;
else
Applied hash table optimisation to RemoveAnchorByNode. This function now takes the anchor name as a parameter, so it can look in the correct bin. In the case of FreeAttrs, we have the name already (since we found a name or id attribute). In the case of FixAnchors, the anchor name could come from either the name or id attribute, so we call the function separately for each case, passing the appropriate attribute value.
2012-08-20 03:06:30 +00:00
attribs->anchor_hash[h] = curr->next;
new
2011-11-17 02:44:16 +00:00
delme = curr;
break;
}
prev = curr;
}
FreeAnchor( doc, delme );
}
/* initialize new anchor */
static Anchor* NewAnchor( TidyDocImpl* doc, ctmbstr name, Node* node )
{
Anchor *a = (Anchor*) TidyDocAlloc( doc, sizeof(Anchor) );
a->name = TY_(tmbstrdup)( doc->allocator, name );
a->name = TY_(tmbstrtolower)(a->name);
a->node = node;
a->next = NULL;
return a;
}
Issue #185 - Treat elements ids as case-sensitive if in HTML5 mode
2015-03-13 18:47:28 +00:00
/*\
* add new anchor to namespace
* Issue #185 - Treat elements ids as case-sensitive
* if in HTML5 modes, make hash of value AS IS!
\*/
new
2011-11-17 02:44:16 +00:00
static Anchor* AddAnchor( TidyDocImpl* doc, ctmbstr name, Node *node )
{
TidyAttribImpl* attribs = &doc->attribs;
Anchor *a = NewAnchor( doc, name, node );
Issue #185 - Treat elements ids as case-sensitive if in HTML5 mode
2015-03-13 18:47:28 +00:00
uint h;
if (TY_(HTMLVersion)(doc) == HT50)
h = anchorNameHash5(name);
else
h = anchorNameHash(name);
new
2011-11-17 02:44:16 +00:00
Use a hash table for anchors
2012-08-19 17:29:16 +00:00
if ( attribs->anchor_hash[h] == NULL)
attribs->anchor_hash[h] = a;
new
2011-11-17 02:44:16 +00:00
else
{
Use a hash table for anchors
2012-08-19 17:29:16 +00:00
Anchor *here = attribs->anchor_hash[h];
new
2011-11-17 02:44:16 +00:00
while (here->next)
here = here->next;
here->next = a;
}
Use a hash table for anchors
2012-08-19 17:29:16 +00:00
return attribs->anchor_hash[h];
new
2011-11-17 02:44:16 +00:00
}
Issue #185 - Treat elements ids as case-sensitive if in HTML5 mode
2015-03-13 18:47:28 +00:00
/*\
* return node associated with anchor
* Issue #185 - Treat elements ids as case-sensitive
* if in HTML5 modes, make hash of value AS IS!
\*/
new
2011-11-17 02:44:16 +00:00
static Node* GetNodeByAnchor( TidyDocImpl* doc, ctmbstr name )
{
TidyAttribImpl* attribs = &doc->attribs;
Anchor *found;
Issue #185 - Treat elements ids as case-sensitive if in HTML5 mode
2015-03-13 18:47:28 +00:00
uint h;
new
2011-11-17 02:44:16 +00:00
tmbstr lname = TY_(tmbstrdup)(doc->allocator, name);
Issue #185 - Treat elements ids as case-sensitive if in HTML5 mode
2015-03-13 18:47:28 +00:00
if (TY_(HTMLVersion)(doc) == HT50) {
h = anchorNameHash5(name);
}
else
{
h = anchorNameHash(name);
lname = TY_(tmbstrtolower)(lname);
}
new
2011-11-17 02:44:16 +00:00
Use a hash table for anchors
2012-08-19 17:29:16 +00:00
for ( found = attribs->anchor_hash[h]; found != NULL; found = found->next )
new
2011-11-17 02:44:16 +00:00
{
if ( TY_(tmbstrcmp)(found->name, lname) == 0 )
break;
}
TidyDocFree(doc, lname);
if ( found )
return found->node;
return NULL;
}
/* free all anchors */
void TY_(FreeAnchors)( TidyDocImpl* doc )
{
TidyAttribImpl* attribs = &doc->attribs;
Anchor* a;
Use a hash table for anchors
2012-08-19 17:29:16 +00:00
uint h;
for (h = 0; h < ANCHOR_HASH_SIZE; h++) {
while (NULL != (a = attribs->anchor_hash[h]) )
{
attribs->anchor_hash[h] = a->next;
FreeAnchor(doc, a);
}
new
2011-11-17 02:44:16 +00:00
}
}
/* public method for inititializing attribute dictionary */
void TY_(InitAttrs)( TidyDocImpl* doc )
{
TidyClearMemory( &doc->attribs, sizeof(TidyAttribImpl) );
#ifdef _DEBUG
{
/* Attribute ID is index position in Attribute type lookup table */
uint ix;
for ( ix=0; ix < N_TIDY_ATTRIBS; ++ix )
{
const Attribute* dict = &attribute_defs[ ix ];
assert( (uint) dict->id == ix );
}
}
#endif
}
/* free all declared attributes */
static void FreeDeclaredAttributes( TidyDocImpl* doc )
{
TidyAttribImpl* attribs = &doc->attribs;
Attribute* dict;
while ( NULL != (dict = attribs->declared_attr_list) )
{
attribs->declared_attr_list = dict->next;
attrsRemoveFromHash( doc, &doc->attribs, dict->name );
TidyDocFree( doc, dict->name );
TidyDocFree( doc, dict );
}
}
void TY_(FreeAttrTable)( TidyDocImpl* doc )
{
attrsEmptyHash( doc, &doc->attribs );
TY_(FreeAnchors)( doc );
FreeDeclaredAttributes( doc );
}
void TY_(AppendToClassAttr)( TidyDocImpl* doc, AttVal *classattr, ctmbstr classname )
{
uint len = TY_(tmbstrlen)(classattr->value) +
TY_(tmbstrlen)(classname) + 2;
tmbstr s = (tmbstr) TidyDocAlloc( doc, len );
s[0] = '\0';
if (classattr->value)
{
TY_(tmbstrcpy)( s, classattr->value );
TY_(tmbstrcat)( s, " " );
}
TY_(tmbstrcat)( s, classname );
if (classattr->value)
TidyDocFree( doc, classattr->value );
classattr->value = s;
}
/* concatenate styles */
static void AppendToStyleAttr( TidyDocImpl* doc, AttVal *styleattr, ctmbstr styleprop )
{
/*
this doesn't handle CSS comments and
leading/trailing white-space very well
see http://www.w3.org/TR/css-style-attr
*/
uint end = TY_(tmbstrlen)(styleattr->value);
if (end >0 && styleattr->value[end - 1] == ';')
{
/* attribute ends with declaration seperator */
styleattr->value = (tmbstr) TidyDocRealloc(doc, styleattr->value,
end + TY_(tmbstrlen)(styleprop) + 2);
TY_(tmbstrcat)(styleattr->value, " ");
TY_(tmbstrcat)(styleattr->value, styleprop);
}
else if (end >0 && styleattr->value[end - 1] == '}')
{
/* attribute ends with rule set */
styleattr->value = (tmbstr) TidyDocRealloc(doc, styleattr->value,
end + TY_(tmbstrlen)(styleprop) + 6);
TY_(tmbstrcat)(styleattr->value, " { ");
TY_(tmbstrcat)(styleattr->value, styleprop);
TY_(tmbstrcat)(styleattr->value, " }");
}
else
{
/* attribute ends with property value */
styleattr->value = (tmbstr) TidyDocRealloc(doc, styleattr->value,
end + TY_(tmbstrlen)(styleprop) + 3);
if (end > 0)
TY_(tmbstrcat)(styleattr->value, "; ");
TY_(tmbstrcat)(styleattr->value, styleprop);
}
}
/*
the same attribute name can't be used
more than once in each element
*/
static Bool AttrsHaveSameName( AttVal* av1, AttVal* av2)
{
TidyAttrId id1, id2;
id1 = AttrId(av1);
id2 = AttrId(av2);
if (id1 != TidyAttr_UNKNOWN && id2 != TidyAttr_UNKNOWN)
return AttrsHaveSameId(av1, av2);
if (id1 != TidyAttr_UNKNOWN || id2 != TidyAttr_UNKNOWN)
return no;
if (av1->attribute && av2->attribute)
return TY_(tmbstrcmp)(av1->attribute, av2->attribute) == 0;
return no;
}
void TY_(RepairDuplicateAttributes)( TidyDocImpl* doc, Node *node, Bool isXml )
{
AttVal *first;
for (first = node->attributes; first != NULL;)
{
AttVal *second;
Bool firstRedefined = no;
if (!(first->asp == NULL && first->php == NULL))
{
first = first->next;
continue;
}
for (second = first->next; second != NULL;)
{
AttVal *temp;
if (!(second->asp == NULL && second->php == NULL
&& AttrsHaveSameName(first, second)))
{
second = second->next;
continue;
}
/* first and second attribute have same local name */
/* now determine what to do with this duplicate... */
if (!isXml
&& attrIsCLASS(first) && cfgBool(doc, TidyJoinClasses)
&& AttrHasValue(first) && AttrHasValue(second))
{
/* concatenate classes */
TY_(AppendToClassAttr)(doc, first, second->value);
temp = second->next;
TY_(ReportAttrError)( doc, node, second, JOINING_ATTRIBUTE);
TY_(RemoveAttribute)( doc, node, second );
second = temp;
}
else if (!isXml
&& attrIsSTYLE(first) && cfgBool(doc, TidyJoinStyles)
&& AttrHasValue(first) && AttrHasValue(second))
{
AppendToStyleAttr( doc, first, second->value );
temp = second->next;
TY_(ReportAttrError)( doc, node, second, JOINING_ATTRIBUTE);
TY_(RemoveAttribute)( doc, node, second );
second = temp;
}
else if ( cfg(doc, TidyDuplicateAttrs) == TidyKeepLast )
{
temp = first->next;
TY_(ReportAttrError)( doc, node, first, REPEATED_ATTRIBUTE);
TY_(RemoveAttribute)( doc, node, first );
firstRedefined = yes;
first = temp;
second = second->next;
}
else /* TidyDuplicateAttrs == TidyKeepFirst */
{
temp = second->next;
TY_(ReportAttrError)( doc, node, second, REPEATED_ATTRIBUTE);
TY_(RemoveAttribute)( doc, node, second );
second = temp;
}
}
if (!firstRedefined)
first = first->next;
}
}
/* ignore unknown attributes for proprietary elements */
const Attribute* TY_(CheckAttribute)( TidyDocImpl* doc, Node *node, AttVal *attval )
{
const Attribute* attribute = attval->dict;
if ( attribute != NULL )
{
Merge in TidyAttr_XML_LANG change.
2012-03-16 11:55:59 +00:00
if (attrIsXML_LANG(attval) || attrIsXML_SPACE(attval))
new
2011-11-17 02:44:16 +00:00
{
doc->lexer->isvoyager = yes;
if (!cfgBool(doc, TidyHtmlOut))
{
TY_(SetOptionBool)(doc, TidyXhtmlOut, yes);
TY_(SetOptionBool)(doc, TidyXmlOut, yes);
}
}
TY_(ConstrainVersion)(doc, AttributeVersions(node, attval));
if (attribute->attrchk)
attribute->attrchk( doc, node, attval );
}
return attribute;
}
Bool TY_(IsBoolAttribute)(AttVal *attval)
{
const Attribute *attribute = ( attval ? attval->dict : NULL );
if ( attribute && attribute->attrchk == CH_BOOL )
return yes;
return no;
}
Bool TY_(attrIsEvent)( AttVal* attval )
{
TidyAttrId atid = AttrId( attval );
return (atid == TidyAttr_OnAFTERUPDATE ||
atid == TidyAttr_OnBEFOREUNLOAD ||
atid == TidyAttr_OnBEFOREUPDATE ||
atid == TidyAttr_OnBLUR ||
atid == TidyAttr_OnCHANGE ||
atid == TidyAttr_OnCLICK ||
atid == TidyAttr_OnDATAAVAILABLE ||
atid == TidyAttr_OnDATASETCHANGED ||
atid == TidyAttr_OnDATASETCOMPLETE ||
atid == TidyAttr_OnDBLCLICK ||
atid == TidyAttr_OnERRORUPDATE ||
atid == TidyAttr_OnFOCUS ||
atid == TidyAttr_OnKEYDOWN ||
atid == TidyAttr_OnKEYPRESS ||
atid == TidyAttr_OnKEYUP ||
atid == TidyAttr_OnLOAD ||
atid == TidyAttr_OnMOUSEDOWN ||
atid == TidyAttr_OnMOUSEMOVE ||
atid == TidyAttr_OnMOUSEOUT ||
atid == TidyAttr_OnMOUSEOVER ||
atid == TidyAttr_OnMOUSEUP ||
atid == TidyAttr_OnRESET ||
atid == TidyAttr_OnROWENTER ||
atid == TidyAttr_OnROWEXIT ||
atid == TidyAttr_OnSELECT ||
atid == TidyAttr_OnSUBMIT ||
atid == TidyAttr_OnUNLOAD);
}
static void CheckLowerCaseAttrValue( TidyDocImpl* doc, Node *node, AttVal *attval)
{
tmbstr p;
Bool hasUpper = no;
if (!AttrHasValue(attval))
return;
p = attval->value;
while (*p)
{
if (TY_(IsUpper)(*p)) /* #501230 - fix by Terry Teague - 09 Jan 02 */
{
hasUpper = yes;
break;
}
p++;
}
if (hasUpper)
{
Lexer* lexer = doc->lexer;
if (lexer->isvoyager)
TY_(ReportAttrError)( doc, node, attval, ATTR_VALUE_NOT_LCASE);
if ( lexer->isvoyager || cfgBool(doc, TidyLowerLiterals) )
attval->value = TY_(tmbstrtolower)(attval->value);
}
}
Fix #588 and #591 - Apply @geoffmcl's patches and tested.
2017-09-29 19:42:03 +00:00
/* Issue #588 - use simple macros only!
Seems 'isalnum(c)' is undefined and can
cause an assert or a SIGSEGV in some libraries
if 'c' is not EOF, or in the range 0 to 0xff,
so avoid using it.
*/
Fix for 'isalnum' in Windows According to the MSN documentation 'isalnum(c)' is only valid when c equals EOF, or is in the range 0 to 255 inclusive. It states the behavior is undefined outside this range, and in Debug mode triggers an assert dialog.
2017-05-08 16:37:04 +00:00
#define ISUPPER(a) ((a >= 'A') && (a <= 'Z'))
#define ISLOWER(a) ((a >= 'a') && (a <= 'z'))
#define ISNUMERIC(a) ((a >= '0') && (a <= '9'))
#define ISALNUM(a) (ISUPPER(a) || ISLOWER(a) || ISNUMERIC(a))
new
2011-11-17 02:44:16 +00:00
Attempt to address issue #352. This patch correctly address the specific issues in #352, but I'm worried that there's some over-reach here. Currently only implemented as a warning, with no switch to turn it off, which maintains current behavior other than the warning. In general, we're treating any string as a complete URL, rather than breaking URL's into component parts. Thus the `IsURLCodePoint()` check includes a few other generic characters that strictly speaking aren't valid codepoints, but are valid as escape characters and delimiters. When addressing #338, I ran into a similar situation in not having a built-in method to separate path components (although a simple generalized solution was good enough in that case). Thus without introducing a new structure and functions to deconstruct a URL into scheme, authority, path, parameters, etc., some variation of this patch will have to be used to address #352.
2017-05-06 22:54:42 +00:00
static Bool IsURLCodePoint( ctmbstr p, uint *increment )
{
uint c;
*increment = TY_(GetUTF8)( p, &c ) + 1;
Fix for 'isalnum' in Windows According to the MSN documentation 'isalnum(c)' is only valid when c equals EOF, or is in the range 0 to 255 inclusive. It states the behavior is undefined outside this range, and in Debug mode triggers an assert dialog.
2017-05-08 16:37:04 +00:00
return ISALNUM( c ) ||
Attempt to address issue #352. This patch correctly address the specific issues in #352, but I'm worried that there's some over-reach here. Currently only implemented as a warning, with no switch to turn it off, which maintains current behavior other than the warning. In general, we're treating any string as a complete URL, rather than breaking URL's into component parts. Thus the `IsURLCodePoint()` check includes a few other generic characters that strictly speaking aren't valid codepoints, but are valid as escape characters and delimiters. When addressing #338, I ran into a similar situation in not having a built-in method to separate path components (although a simple generalized solution was good enough in that case). Thus without introducing a new structure and functions to deconstruct a URL into scheme, authority, path, parameters, etc., some variation of this patch will have to be used to address #352.
2017-05-06 22:54:42 +00:00
c == '%' || /* not a valid codepoint, but an escape sequence */
c == '#' || /* not a valid codepoint, but a delimiter */
c == '!' ||
c == '$' ||
c == '&' ||
c == '\'' ||
c == '(' ||
c == ')' ||
c == '*' ||
c == '+' ||
c == ',' ||
c == '-' ||
c == '.' ||
c == '/' ||
c == ':' ||
c == ';' ||
c == '=' ||
c == '?' ||
c == '@' ||
c == '_' ||
c == '~' ||
(c >= 0x00A0 && c <= 0xD7FF) ||
(c >= 0xE000 && c <= 0xFDCF) ||
(c >= 0xFDF0 && c <= 0xFFEF) ||
(c >= 0x10000 && c <= 0x1FFFD) ||
(c >= 0x20000 && c <= 0x2FFFD) ||
(c >= 0x30000 && c <= 0x3FFFD) ||
(c >= 0x40000 && c <= 0x4FFFD) ||
(c >= 0x50000 && c <= 0x5FFFD) ||
(c >= 0x60000 && c <= 0x6FFFD) ||
(c >= 0x70000 && c <= 0x7FFFD) ||
(c >= 0x80000 && c <= 0x8FFFD) ||
(c >= 0x90000 && c <= 0x9FFFD) ||
(c >= 0xA0000 && c <= 0xAFFFD) ||
(c >= 0xB0000 && c <= 0xBFFFD) ||
(c >= 0xC0000 && c <= 0xCFFFD) ||
(c >= 0xD0000 && c <= 0xDFFFD) ||
(c >= 0xE0000 && c <= 0xEFFFD) ||
(c >= 0xF0000 && c <= 0xFFFFD) ||
(c >= 0x100000 && c <= 0x10FFFD);
}
new
2011-11-17 02:44:16 +00:00
void TY_(CheckUrl)( TidyDocImpl* doc, Node *node, AttVal *attval)
{
Attempt to address issue #352. This patch correctly address the specific issues in #352, but I'm worried that there's some over-reach here. Currently only implemented as a warning, with no switch to turn it off, which maintains current behavior other than the warning. In general, we're treating any string as a complete URL, rather than breaking URL's into component parts. Thus the `IsURLCodePoint()` check includes a few other generic characters that strictly speaking aren't valid codepoints, but are valid as escape characters and delimiters. When addressing #338, I ran into a similar situation in not having a built-in method to separate path components (although a simple generalized solution was good enough in that case). Thus without introducing a new structure and functions to deconstruct a URL into scheme, authority, path, parameters, etc., some variation of this patch will have to be used to address #352.
2017-05-06 22:54:42 +00:00
tmbchar c;
new
2011-11-17 02:44:16 +00:00
tmbstr dest, p;
Attempt to address issue #352. This patch correctly address the specific issues in #352, but I'm worried that there's some over-reach here. Currently only implemented as a warning, with no switch to turn it off, which maintains current behavior other than the warning. In general, we're treating any string as a complete URL, rather than breaking URL's into component parts. Thus the `IsURLCodePoint()` check includes a few other generic characters that strictly speaking aren't valid codepoints, but are valid as escape characters and delimiters. When addressing #338, I ran into a similar situation in not having a built-in method to separate path components (although a simple generalized solution was good enough in that case). Thus without introducing a new structure and functions to deconstruct a URL into scheme, authority, path, parameters, etc., some variation of this patch will have to be used to address #352.
2017-05-06 22:54:42 +00:00
uint escape_count = 0, backslash_count = 0, bad_codepoint_count = 0;
new
2011-11-17 02:44:16 +00:00
uint i, pos = 0;
uint len;
Attempt to address issue #352. This patch correctly address the specific issues in #352, but I'm worried that there's some over-reach here. Currently only implemented as a warning, with no switch to turn it off, which maintains current behavior other than the warning. In general, we're treating any string as a complete URL, rather than breaking URL's into component parts. Thus the `IsURLCodePoint()` check includes a few other generic characters that strictly speaking aren't valid codepoints, but are valid as escape characters and delimiters. When addressing #338, I ran into a similar situation in not having a built-in method to separate path components (although a simple generalized solution was good enough in that case). Thus without introducing a new structure and functions to deconstruct a URL into scheme, authority, path, parameters, etc., some variation of this patch will have to be used to address #352.
2017-05-06 22:54:42 +00:00
uint increment;
new
2011-11-17 02:44:16 +00:00
Bool isJavascript = no;
Attempt to address issue #352. This patch correctly address the specific issues in #352, but I'm worried that there's some over-reach here. Currently only implemented as a warning, with no switch to turn it off, which maintains current behavior other than the warning. In general, we're treating any string as a complete URL, rather than breaking URL's into component parts. Thus the `IsURLCodePoint()` check includes a few other generic characters that strictly speaking aren't valid codepoints, but are valid as escape characters and delimiters. When addressing #338, I ran into a similar situation in not having a built-in method to separate path components (although a simple generalized solution was good enough in that case). Thus without introducing a new structure and functions to deconstruct a URL into scheme, authority, path, parameters, etc., some variation of this patch will have to be used to address #352.
2017-05-06 22:54:42 +00:00
new
2011-11-17 02:44:16 +00:00
if (!AttrHasValue(attval))
{
TY_(ReportAttrError)( doc, node, attval, MISSING_ATTR_VALUE);
return;
}
p = attval->value;
Attempt to address issue #352. This patch correctly address the specific issues in #352, but I'm worried that there's some over-reach here. Currently only implemented as a warning, with no switch to turn it off, which maintains current behavior other than the warning. In general, we're treating any string as a complete URL, rather than breaking URL's into component parts. Thus the `IsURLCodePoint()` check includes a few other generic characters that strictly speaking aren't valid codepoints, but are valid as escape characters and delimiters. When addressing #338, I ran into a similar situation in not having a built-in method to separate path components (although a simple generalized solution was good enough in that case). Thus without introducing a new structure and functions to deconstruct a URL into scheme, authority, path, parameters, etc., some variation of this patch will have to be used to address #352.
2017-05-06 22:54:42 +00:00
new
2011-11-17 02:44:16 +00:00
isJavascript =
TY_(tmbstrncmp)(p,"javascript:",sizeof("javascript:")-1)==0;
for (i = 0; '\0' != (c = p[i]); ++i)
{
if (c == '\\')
{
++backslash_count;
if ( cfgBool(doc, TidyFixBackslash) && !isJavascript)
p[i] = '/';
}
else if ((c > 0x7e) || (c <= 0x20) || (strchr("<>", c)))
++escape_count;
}
Issue #345 - discard leading spaces in href
2016-02-01 19:07:55 +00:00
Attempt to address issue #352. This patch correctly address the specific issues in #352, but I'm worried that there's some over-reach here. Currently only implemented as a warning, with no switch to turn it off, which maintains current behavior other than the warning. In general, we're treating any string as a complete URL, rather than breaking URL's into component parts. Thus the `IsURLCodePoint()` check includes a few other generic characters that strictly speaking aren't valid codepoints, but are valid as escape characters and delimiters. When addressing #338, I ran into a similar situation in not having a built-in method to separate path components (although a simple generalized solution was good enough in that case). Thus without introducing a new structure and functions to deconstruct a URL into scheme, authority, path, parameters, etc., some variation of this patch will have to be used to address #352.
2017-05-06 22:54:42 +00:00
while ( *p != 0 )
{
if ( !IsURLCodePoint( p, &increment ) )
++bad_codepoint_count;
p = p + increment;
}
p = attval->value;
new
2011-11-17 02:44:16 +00:00
if ( cfgBool(doc, TidyFixUri) && escape_count )
{
Issue #345 - discard leading spaces in href
2016-02-01 19:07:55 +00:00
Bool hadnonspace = no;
new
2011-11-17 02:44:16 +00:00
len = TY_(tmbstrlen)(p) + escape_count * 2 + 1;
dest = (tmbstr) TidyDocAlloc(doc, len);
Issue #345 - discard leading spaces in href
2016-02-01 19:07:55 +00:00
new
2011-11-17 02:44:16 +00:00
for (i = 0; 0 != (c = p[i]); ++i)
{
if ((c > 0x7e) || (c <= 0x20) || (strchr("<>", c)))
Issue #345 - discard leading spaces in href
2016-02-01 19:07:55 +00:00
{
if (c == 0x20)
{
/* #345 - special case for leading spaces - discard */
if (hadnonspace)
pos += sprintf( dest + pos, "%%%02X", (byte)c );
}
else
{
pos += sprintf( dest + pos, "%%%02X", (byte)c );
hadnonspace = yes;
}
}
new
2011-11-17 02:44:16 +00:00
else
Issue #345 - discard leading spaces in href
2016-02-01 19:07:55 +00:00
{
hadnonspace = yes;
new
2011-11-17 02:44:16 +00:00
dest[pos++] = c;
Issue #345 - discard leading spaces in href
2016-02-01 19:07:55 +00:00
}
new
2011-11-17 02:44:16 +00:00
}
dest[pos] = 0;
TidyDocFree(doc, attval->value);
attval->value = dest;
}
if ( backslash_count )
{
if ( cfgBool(doc, TidyFixBackslash) && !isJavascript )
TY_(ReportAttrError)( doc, node, attval, FIXED_BACKSLASH );
else
TY_(ReportAttrError)( doc, node, attval, BACKSLASH_IN_URI );
}
if ( escape_count )
{
if ( cfgBool(doc, TidyFixUri) )
TY_(ReportAttrError)( doc, node, attval, ESCAPED_ILLEGAL_URI);
Address #378 Addresses issue #378 by NOT emitting warnings if `fix-uri` is `no`, for HTML5 documents. This preserves existing behavior for legacy document types.
2017-05-14 00:46:48 +00:00
else if ( !(TY_(HTMLVersion)(doc) & VERS_HTML5) )
new
2011-11-17 02:44:16 +00:00
TY_(ReportAttrError)( doc, node, attval, ILLEGAL_URI_REFERENCE);
doc->badChars |= BC_INVALID_URI;
}
Attempt to address issue #352. This patch correctly address the specific issues in #352, but I'm worried that there's some over-reach here. Currently only implemented as a warning, with no switch to turn it off, which maintains current behavior other than the warning. In general, we're treating any string as a complete URL, rather than breaking URL's into component parts. Thus the `IsURLCodePoint()` check includes a few other generic characters that strictly speaking aren't valid codepoints, but are valid as escape characters and delimiters. When addressing #338, I ran into a similar situation in not having a built-in method to separate path components (although a simple generalized solution was good enough in that case). Thus without introducing a new structure and functions to deconstruct a URL into scheme, authority, path, parameters, etc., some variation of this patch will have to be used to address #352.
2017-05-06 22:54:42 +00:00
if ( bad_codepoint_count )
{
TY_(ReportAttrError)( doc, node, attval, ILLEGAL_URI_CODEPOINT );
}
new
2011-11-17 02:44:16 +00:00
}
/* RFC 2396, section 4.2 states:
"[...] in the case of HTML's FORM element, [...] an
empty URI reference represents the base URI of the
current document and should be replaced by that URI
when transformed into a request."
*/
void CheckAction( TidyDocImpl* doc, Node *node, AttVal *attval)
{
if (AttrHasValue(attval))
TY_(CheckUrl)( doc, node, attval );
}
void CheckScript( TidyDocImpl* ARG_UNUSED(doc), Node* ARG_UNUSED(node),
AttVal* ARG_UNUSED(attval))
{
}
Bool TY_(IsValidHTMLID)(ctmbstr id)
{
ctmbstr s = id;
if (!s)
return no;
while (*s)
HTML IDs can contain anything except whitespace. Introduced TY_(IsHTMLSpace)(uint c), which checks to see if c is one of the chars that the HTML spec (and browsers) treat as a space in attribute values: 0x020 (space), 0x009 (tab), 0x00a (LF), 0x00c (FF), or 0x00d (CF). Can't use ANSI C isspace(int c) here because like standard functions for many other langs, it also treats 0x00b as a space.
2012-01-02 07:12:51 +00:00
if (TY_(IsHTMLSpace)(*s++))
new
2011-11-17 02:44:16 +00:00
return no;
return yes;
}
Bool TY_(IsValidXMLID)(ctmbstr id)
{
ctmbstr s = id;
tchar c;
if (!s)
return no;
c = *s++;
if (c > 0x7F)
s += TY_(GetUTF8)(s, &c);
if (!(TY_(IsXMLLetter)(c) || c == '_' || c == ':'))
return no;
while (*s)
{
c = (unsigned char)*s;
if (c > 0x7F)
s += TY_(GetUTF8)(s, &c);
++s;
if (!TY_(IsXMLNamechar)(c))
return no;
}
return yes;
}
static Bool IsValidNMTOKEN(ctmbstr name)
{
ctmbstr s = name;
tchar c;
if (!s)
return no;
while (*s)
{
c = (unsigned char)*s;
if (c > 0x7F)
s += TY_(GetUTF8)(s, &c);
++s;
if (!TY_(IsXMLNamechar)(c))
return no;
}
return yes;
}
static Bool AttrValueIsAmong(AttVal *attval, ctmbstr const list[])
{
const ctmbstr *v;
for (v = list; *v; ++v)
if (AttrValueIs(attval, *v))
return yes;
return no;
}
static void CheckAttrValidity( TidyDocImpl* doc, Node *node, AttVal *attval,
ctmbstr const list[])
{
if (!AttrHasValue(attval))
{
TY_(ReportAttrError)( doc, node, attval, MISSING_ATTR_VALUE);
return;
}
CheckLowerCaseAttrValue( doc, node, attval );
if (!AttrValueIsAmong(attval, list))
TY_(ReportAttrError)( doc, node, attval, BAD_ATTRIBUTE_VALUE);
}
void CheckName( TidyDocImpl* doc, Node *node, AttVal *attval)
{
Node *old;
if (!AttrHasValue(attval))
{
TY_(ReportAttrError)( doc, node, attval, MISSING_ATTR_VALUE);
return;
}
if ( TY_(IsAnchorElement)(doc, node) )
{
if (cfgBool(doc, TidyXmlOut) && !IsValidNMTOKEN(attval->value))
TY_(ReportAttrError)( doc, node, attval, BAD_ATTRIBUTE_VALUE);
if ((old = GetNodeByAnchor(doc, attval->value)) && old != node)
{
TY_(ReportAttrError)( doc, node, attval, ANCHOR_NOT_UNIQUE);
}
else
AddAnchor( doc, attval->value, node );
}
}
void CheckId( TidyDocImpl* doc, Node *node, AttVal *attval )
{
Lexer* lexer = doc->lexer;
Node *old;
if (!AttrHasValue(attval))
{
TY_(ReportAttrError)( doc, node, attval, MISSING_ATTR_VALUE);
return;
}
if (!TY_(IsValidHTMLID)(attval->value))
{
if (lexer->isvoyager && TY_(IsValidXMLID)(attval->value))
TY_(ReportAttrError)( doc, node, attval, XML_ID_SYNTAX);
else
TY_(ReportAttrError)( doc, node, attval, BAD_ATTRIBUTE_VALUE);
}
if ((old = GetNodeByAnchor(doc, attval->value)) && old != node)
{
TY_(ReportAttrError)( doc, node, attval, ANCHOR_NOT_UNIQUE);
}
else
AddAnchor( doc, attval->value, node );
}
- Add support for the `is` attribute. - Add support for autonomous custom elements.
2017-03-13 17:45:32 +00:00
void CheckIs( TidyDocImpl* doc, Node *node, AttVal *attval )
{
const char *ptr;
Bool go = yes;
/* `is` MUST NOT be in an autonomous custom tag */
ptr = strchr(node->element, '-');
if ( ( ptr && (ptr - node->element > 0) ) )
{
TY_(ReportAttrError)( doc, node, attval, ATTRIBUTE_IS_NOT_ALLOWED);
}
/* Even if we fail the above test, we'll continue to emit reports because
the user should *also* know that his attribute values are wrong, even
if they should't be in custom tags anyway. */
/* `is` MUST have a value */
if (!AttrHasValue(attval))
{
TY_(ReportAttrError)( doc, node, attval, MISSING_ATTR_VALUE);
return;
}
/* `is` MUST contain a hyphen and no space. */
ptr = strchr(attval->value, '-');
go = ( ptr && (ptr - attval->value > 0) );
ptr = strchr(attval->value, ' ');
go = go & (ptr == NULL);
if ( !go )
{
TY_(ReportAttrError)( doc, node, attval, BAD_ATTRIBUTE_VALUE);
}
}
new
2011-11-17 02:44:16 +00:00
void CheckBool( TidyDocImpl* doc, Node *node, AttVal *attval)
{
if (!AttrHasValue(attval))
return;
CheckLowerCaseAttrValue( doc, node, attval );
}
void CheckAlign( TidyDocImpl* doc, Node *node, AttVal *attval)
{
ctmbstr const values[] = {"left", "right", "center", "justify", NULL};
/* IMG, OBJECT, APPLET and EMBED use align for vertical position */
if (node->tag && (node->tag->model & CM_IMG))
{
CheckValign( doc, node, attval );
return;
}
if (!AttrHasValue(attval))
{
TY_(ReportAttrError)( doc, node, attval, MISSING_ATTR_VALUE);
return;
}
CheckLowerCaseAttrValue( doc, node, attval);
/* currently CheckCaption(...) takes care of the remaining cases */
if (nodeIsCAPTION(node))
return;
if (!AttrValueIsAmong(attval, values))
{
/* align="char" is allowed for elements with CM_TABLE|CM_ROW
except CAPTION which is excluded above, */
if( !(AttrValueIs(attval, "char")
&& TY_(nodeHasCM)(node, CM_TABLE|CM_ROW)) )
TY_(ReportAttrError)( doc, node, attval, BAD_ATTRIBUTE_VALUE);
}
}
void CheckValign( TidyDocImpl* doc, Node *node, AttVal *attval)
{
ctmbstr const values[] = {"top", "middle", "bottom", "baseline", NULL};
ctmbstr const values2[] = {"left", "right", NULL};
ctmbstr const valuesp[] = {"texttop", "absmiddle", "absbottom",
"textbottom", NULL};
if (!AttrHasValue(attval))
{
TY_(ReportAttrError)( doc, node, attval, MISSING_ATTR_VALUE);
return;
}
CheckLowerCaseAttrValue( doc, node, attval );
if (AttrValueIsAmong(attval, values))
{
/* all is fine */
}
else if (AttrValueIsAmong(attval, values2))
{
if (!(node->tag && (node->tag->model & CM_IMG)))
TY_(ReportAttrError)( doc, node, attval, BAD_ATTRIBUTE_VALUE);
}
else if (AttrValueIsAmong(attval, valuesp))
{
TY_(ConstrainVersion)( doc, VERS_PROPRIETARY );
TY_(ReportAttrError)( doc, node, attval, PROPRIETARY_ATTR_VALUE);
}
else
TY_(ReportAttrError)( doc, node, attval, BAD_ATTRIBUTE_VALUE);
}
void CheckLength( TidyDocImpl* doc, Node *node, AttVal *attval)
{
tmbstr p;
if (!AttrHasValue(attval))
{
TY_(ReportAttrError)( doc, node, attval, MISSING_ATTR_VALUE);
return;
}
/* don't check for <col width=...> and <colgroup width=...> */
if (attrIsWIDTH(attval) && (nodeIsCOL(node) || nodeIsCOLGROUP(node)))
return;
p = attval->value;
if (!TY_(IsDigit)(*p++))
{
TY_(ReportAttrError)( doc, node, attval, BAD_ATTRIBUTE_VALUE);
}
else
{
while (*p)
{
if (!TY_(IsDigit)(*p) && *p != '%')
{
TY_(ReportAttrError)( doc, node, attval, BAD_ATTRIBUTE_VALUE);
break;
}
++p;
}
}
}
void CheckTarget( TidyDocImpl* doc, Node *node, AttVal *attval)
{
ctmbstr const values[] = {"_blank", "_self", "_parent", "_top", NULL};
if (!AttrHasValue(attval))
{
TY_(ReportAttrError)( doc, node, attval, MISSING_ATTR_VALUE);
return;
}
/* target names must begin with A-Za-z ... */
if (TY_(IsLetter)(attval->value[0]))
return;
/* or be one of the allowed list */
if (!AttrValueIsAmong(attval, values))
TY_(ReportAttrError)( doc, node, attval, BAD_ATTRIBUTE_VALUE);
}
void CheckFsubmit( TidyDocImpl* doc, Node *node, AttVal *attval)
{
ctmbstr const values[] = {"get", "post", NULL};
CheckAttrValidity( doc, node, attval, values );
}
void CheckClear( TidyDocImpl* doc, Node *node, AttVal *attval)
{
ctmbstr const values[] = {"none", "left", "right", "all", NULL};
if (!AttrHasValue(attval))
{
TY_(ReportAttrError)( doc, node, attval, MISSING_ATTR_VALUE);
if (attval->value == NULL)
attval->value = TY_(tmbstrdup)( doc->allocator, "none" );
return;
}
CheckLowerCaseAttrValue( doc, node, attval );
if (!AttrValueIsAmong(attval, values))
TY_(ReportAttrError)( doc, node, attval, BAD_ATTRIBUTE_VALUE);
}
void CheckShape( TidyDocImpl* doc, Node *node, AttVal *attval)
{
ctmbstr const values[] = {"rect", "default", "circle", "poly", NULL};
CheckAttrValidity( doc, node, attval, values );
}
void CheckScope( TidyDocImpl* doc, Node *node, AttVal *attval)
{
ctmbstr const values[] = {"row", "rowgroup", "col", "colgroup", NULL};
CheckAttrValidity( doc, node, attval, values );
}
void CheckNumber( TidyDocImpl* doc, Node *node, AttVal *attval)
{
tmbstr p;
if (!AttrHasValue(attval))
{
TY_(ReportAttrError)( doc, node, attval, MISSING_ATTR_VALUE);
return;
}
/* don't check <frameset cols=... rows=...> */
if ( nodeIsFRAMESET(node) &&
(attrIsCOLS(attval) || attrIsROWS(attval)))
return;
p = attval->value;
/* font size may be preceded by + or - */
if ( nodeIsFONT(node) && (*p == '+' || *p == '-') )
++p;
exception attr tabindex can begin with '-'
2015-01-18 13:46:12 +00:00
/* tabindex may be preceeded by - */
if (attval->attribute && (strcmp(attval->attribute,"tabindex") == 0) && (*p == '-'))
++p;
new
2011-11-17 02:44:16 +00:00
while (*p)
{
if (!TY_(IsDigit)(*p))
{
TY_(ReportAttrError)( doc, node, attval, BAD_ATTRIBUTE_VALUE);
break;
}
++p;
}
}
/* check hexadecimal color value */
static Bool IsValidColorCode(ctmbstr color)
{
uint i;
if (TY_(tmbstrlen)(color) != 6)
return no;
/* check if valid hex digits and letters */
for (i = 0; i < 6; i++)
if (!TY_(IsDigit)(color[i]) && !strchr("abcdef", TY_(ToLower)(color[i])))
return no;
return yes;
}
/* check color syntax and beautify value by option */
void CheckColor( TidyDocImpl* doc, Node *node, AttVal *attval)
{
Bool valid = no;
tmbstr given;
if (!AttrHasValue(attval))
{
TY_(ReportAttrError)( doc, node, attval, MISSING_ATTR_VALUE);
return;
}
given = attval->value;
/* 727851 - add hash to hash-less color values */
if (given[0] != '#' && (valid = IsValidColorCode(given)))
{
tmbstr cp, s;
cp = s = (tmbstr) TidyDocAlloc(doc, 2 + TY_(tmbstrlen)(given));
*cp++ = '#';
while ('\0' != (*cp++ = *given++))
continue;
TY_(ReportAttrError)(doc, node, attval, BAD_ATTRIBUTE_VALUE_REPLACED);
TidyDocFree(doc, attval->value);
given = attval->value = s;
}
if (!valid && given[0] == '#')
valid = IsValidColorCode(given + 1);
if (valid && given[0] == '#' && cfgBool(doc, TidyReplaceColor))
{
ctmbstr newName = GetColorName(given);
if (newName)
{
TidyDocFree(doc, attval->value);
given = attval->value = TY_(tmbstrdup)(doc->allocator, newName);
}
}
/* if it is not a valid color code, it is a color name */
if (!valid)
valid = GetColorCode(given) != NULL;
if (valid && given[0] == '#')
attval->value = TY_(tmbstrtoupper)(attval->value);
else if (valid)
attval->value = TY_(tmbstrtolower)(attval->value);
if (!valid)
TY_(ReportAttrError)( doc, node, attval, BAD_ATTRIBUTE_VALUE);
}
/* check valuetype attribute for element param */
void CheckVType( TidyDocImpl* doc, Node *node, AttVal *attval)
{
ctmbstr const values[] = {"data", "object", "ref", NULL};
CheckAttrValidity( doc, node, attval, values );
}
/* checks scrolling attribute */
void CheckScroll( TidyDocImpl* doc, Node *node, AttVal *attval)
{
ctmbstr const values[] = {"no", "auto", "yes", NULL};
CheckAttrValidity( doc, node, attval, values );
}
/* checks dir attribute */
void CheckTextDir( TidyDocImpl* doc, Node *node, AttVal *attval)
{
PR #712 - Only allow 'auto' on 'dir' for HTML5 modified: src/attrs.c
2018-04-23 17:00:36 +00:00
ctmbstr const values4[] = { "rtl", "ltr", NULL };
/* PR #712 - add 'auto' for HTML5 - @doronbehar */
ctmbstr const values5[] = { "rtl", "ltr", "auto", NULL };
CheckAttrValidity(doc, node, attval,
(TY_(IsHTML5Mode)(doc) ? values5 : values4));
new
2011-11-17 02:44:16 +00:00
}
/* checks lang and xml:lang attributes */
void CheckLang( TidyDocImpl* doc, Node *node, AttVal *attval)
{
/* empty xml:lang is allowed through XML 1.0 SE errata */
if (!AttrHasValue(attval) && !attrIsXML_LANG(attval))
{
if ( cfg(doc, TidyAccessibilityCheckLevel) == 0 )
{
TY_(ReportAttrError)( doc, node, attval, MISSING_ATTR_VALUE );
}
return;
}
}
/* checks type attribute */
void CheckType( TidyDocImpl* doc, Node *node, AttVal *attval)
{
Accept HTML5 input types.
2012-06-07 14:54:04 +00:00
ctmbstr const valuesINPUT[] = {
"text", "password", "checkbox", "radio", "submit", "reset", "file",
"hidden", "image", "button", "color", "date", "datetime",
"datetime-local", "email", "month", "number", "range", "search",
"tel", "time", "url", "week", NULL};
new
2011-11-17 02:44:16 +00:00
ctmbstr const valuesBUTTON[] = {"button", "submit", "reset", NULL};
ctmbstr const valuesUL[] = {"disc", "square", "circle", NULL};
ctmbstr const valuesOL[] = {"1", "a", "i", NULL};
if (nodeIsINPUT(node))
CheckAttrValidity( doc, node, attval, valuesINPUT );
else if (nodeIsBUTTON(node))
CheckAttrValidity( doc, node, attval, valuesBUTTON );
else if (nodeIsUL(node))
CheckAttrValidity( doc, node, attval, valuesUL );
else if (nodeIsOL(node))
{
if (!AttrHasValue(attval))
{
TY_(ReportAttrError)( doc, node, attval, MISSING_ATTR_VALUE);
return;
}
if (!AttrValueIsAmong(attval, valuesOL))
TY_(ReportAttrError)( doc, node, attval, BAD_ATTRIBUTE_VALUE);
}
else if (nodeIsLI(node))
{
if (!AttrHasValue(attval))
{
TY_(ReportAttrError)( doc, node, attval, MISSING_ATTR_VALUE);
return;
}
if (AttrValueIsAmong(attval, valuesUL))
CheckLowerCaseAttrValue( doc, node, attval );
else if (!AttrValueIsAmong(attval, valuesOL))
TY_(ReportAttrError)( doc, node, attval, BAD_ATTRIBUTE_VALUE);
}
return;
}
static
New config option now drives the list; memory deallocated.
2017-09-23 11:20:31 +00:00
AttVal *SortAttVal( TidyDocImpl* doc, AttVal* list, TidyAttrSortStrategy strat );
new
2011-11-17 02:44:16 +00:00
New config option now drives the list; memory deallocated.
2017-09-23 11:20:31 +00:00
void TY_(SortAttributes)(TidyDocImpl* doc, Node* node, TidyAttrSortStrategy strat)
new
2011-11-17 02:44:16 +00:00
{
while (node)
{
New config option now drives the list; memory deallocated.
2017-09-23 11:20:31 +00:00
node->attributes = SortAttVal( doc, node->attributes, strat );
new
2011-11-17 02:44:16 +00:00
if (node->content)
New config option now drives the list; memory deallocated.
2017-09-23 11:20:31 +00:00
TY_(SortAttributes)(doc, node->content, strat);
new
2011-11-17 02:44:16 +00:00
node = node->next;
}
}
/**
* Attribute sorting contributed by Adrian Wilkins, 2007
*
* Portions copyright Simon Tatham 2001.
*
* Merge sort algortithm adpated from listsort.c linked from
* http://www.chiark.greenend.org.uk/~sgtatham/algorithms/listsort.html
*
* Original copyright notice proceeds below.
*
* Permission is hereby granted, free of charge, to any person
* obtaining a copy of this software and associated documentation
* files (the "Software"), to deal in the Software without
* restriction, including without limitation the rights to use,
* copy, modify, merge, publish, distribute, sublicense, and/or
* sell copies of the Software, and to permit persons to whom the
* Software is furnished to do so, subject to the following
* conditions:
*
* The above copyright notice and this permission notice shall be
* included in all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES
* OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
* NONINFRINGEMENT. IN NO EVENT SHALL SIMON TATHAM BE LIABLE FOR
* ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF
* CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
* CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*/
Initial commit of #434; WIP
2017-09-23 02:14:59 +00:00
typedef int(*ptAttValComparator)(AttVal *one, AttVal *two, ctmbstr *list);
/* Returns the index of the item in the array, or -1 if not in the array */
static
int indexof( ctmbstr item, ctmbstr *list )
{
New config option now drives the list; memory deallocated.
2017-09-23 11:20:31 +00:00
if ( list )
{
uint i = 0;
while ( list[i] != NULL ) {
if ( TY_(tmbstrcasecmp)(item, list[i]) == 0 )
return i;
i++;
}
Initial commit of #434; WIP
2017-09-23 02:14:59 +00:00
}
return -1;
}
new
2011-11-17 02:44:16 +00:00
Initial commit of #434; WIP
2017-09-23 02:14:59 +00:00
/* Comparison function for TidySortAttrAlpha. Will also consider items in
the passed in list as higher-priority, and will group them first.
*/
new
2011-11-17 02:44:16 +00:00
static
Initial commit of #434; WIP
2017-09-23 02:14:59 +00:00
int AlphaComparator(AttVal *one, AttVal *two, ctmbstr *list)
new
2011-11-17 02:44:16 +00:00
{
Initial commit of #434; WIP
2017-09-23 02:14:59 +00:00
int oneIndex = indexof( one->attribute, list );
int twoIndex = indexof( two->attribute, list );
/* If both on the list, the lower index has priority. */
if ( oneIndex >= 0 && twoIndex >= 0 )
return oneIndex < twoIndex ? -1 : 1;
/* If A on the list but B not on the list, then A has priority. */
if ( oneIndex >= 0 && twoIndex == -1 )
return -1;
/* If A not on the list but B is on the list, then B has priority. */
if ( oneIndex == -1 && twoIndex >= 0 )
return 1;
/* Otherwise nothing is on the list, so just compare strings. */
new
2011-11-17 02:44:16 +00:00
return TY_(tmbstrcmp)(one->attribute, two->attribute);
}
Initial commit of #434; WIP
2017-09-23 02:14:59 +00:00
/* Comparison function for prioritizing list items. It doesn't otherwise
sort.
*/
static
int PriorityComparator(AttVal *one, AttVal *two, ctmbstr *list)
{
int oneIndex = indexof( one->attribute, list );
int twoIndex = indexof( two->attribute, list );
/* If both on the list, the lower index has priority. */
if ( oneIndex >= 0 && twoIndex >= 0 )
return oneIndex < twoIndex ? -1 : 1;
/* If A on the list but B not on the list, then A has priority. */
if ( oneIndex >= 0 && twoIndex == -1 )
return -1;
/* If A not on the list but B is on the list, then B has priority. */
if ( oneIndex == -1 && twoIndex >= 0 )
return 1;
/* Otherwise nothing is on the list, so just mark them as the same. */
return 0;
}
new
2011-11-17 02:44:16 +00:00
/* The "factory method" that returns a pointer to the comparator function */
static
Initial commit of #434; WIP
2017-09-23 02:14:59 +00:00
ptAttValComparator GetAttValComparator(TidyAttrSortStrategy strat, ctmbstr *list)
new
2011-11-17 02:44:16 +00:00
{
switch (strat)
{
case TidySortAttrAlpha:
return AlphaComparator;
case TidySortAttrNone:
New config option now drives the list; memory deallocated.
2017-09-23 11:20:31 +00:00
if ( list && list[0] )
Initial commit of #434; WIP
2017-09-23 02:14:59 +00:00
return PriorityComparator;
new
2011-11-17 02:44:16 +00:00
break;
}
return 0;
}
/* The sort routine */
static
New config option now drives the list; memory deallocated.
2017-09-23 11:20:31 +00:00
AttVal *SortAttVal( TidyDocImpl* doc, AttVal *list, TidyAttrSortStrategy strat)
new
2011-11-17 02:44:16 +00:00
{
New config option now drives the list; memory deallocated.
2017-09-23 11:20:31 +00:00
/* Get the list from the passed-in tidyDoc. */
ctmbstr* priorityList = (ctmbstr*)doc->attribs.priorityAttribs.list;
Initial commit of #434; WIP
2017-09-23 02:14:59 +00:00
New config option now drives the list; memory deallocated.
2017-09-23 11:20:31 +00:00
ptAttValComparator ptComparator = GetAttValComparator(strat, priorityList);
new
2011-11-17 02:44:16 +00:00
AttVal *p, *q, *e, *tail;
int insize, nmerges, psize, qsize, i;
/*
* Silly special case: if `list' was passed in as NULL, return
* NULL immediately.
*/
if (!list)
return NULL;
New config option now drives the list; memory deallocated.
2017-09-23 11:20:31 +00:00
/* If no comparator, return the list as-is */
Initial commit of #434; WIP
2017-09-23 02:14:59 +00:00
if (ptComparator == 0)
return list;
new
2011-11-17 02:44:16 +00:00
insize = 1;
while (1) {
p = list;
list = NULL;
tail = NULL;
nmerges = 0; /* count number of merges we do in this pass */
while (p) {
nmerges++; /* there exists a merge to be done */
/* step `insize' places along from p */
q = p;
psize = 0;
for (i = 0; i < insize; i++) {
psize++;
q = q->next;
if(!q) break;
}
/* if q hasn't fallen off end, we have two lists to merge */
qsize = insize;
/* now we have two lists; merge them */
while (psize > 0 || (qsize > 0 && q)) {
/* decide whether next element of merge comes from p or q */
if (psize == 0) {
/* p is empty; e must come from q. */
e = q; q = q->next; qsize--;
} else if (qsize == 0 || !q) {
/* q is empty; e must come from p. */
e = p; p = p->next; psize--;
New config option now drives the list; memory deallocated.
2017-09-23 11:20:31 +00:00
} else if (ptComparator(p,q, priorityList) <= 0) {
new
2011-11-17 02:44:16 +00:00
/* First element of p is lower (or same);
* e must come from p. */
e = p; p = p->next; psize--;
} else {
/* First element of q is lower; e must come from q. */
e = q; q = q->next; qsize--;
}
/* add the next element to the merged list */
if (tail) {
tail->next = e;
} else {
list = e;
}
tail = e;
}
/* now p has stepped `insize' places along, and q has too */
p = q;
}
tail->next = NULL;
/* If we have done only one merge, we're finished. */
if (nmerges <= 1) /* allow for nmerges==0, the empty list case */
return list;
/* Otherwise repeat, merging lists twice the size */
insize *= 2;
}
}
Initial cut at RDFa support (again) New branch that implements support for RDFa attributes. Should be cleaner than my first attempt in PR #299 - also references issue #209
2015-11-16 17:29:23 +00:00
/* RDFA support checkers
*
*/
/* CheckRDFAPrefix - ensure the prefix attribute value is
* correct
*
* @prefix takes prefix value pairs in the form:
*
* NCName ':' ' '+ AnyURI
*/
void CheckRDFaPrefix ( TidyDocImpl* doc, Node *node, AttVal *attval)
{
if (!AttrHasValue(attval))
{
TY_(ReportAttrError)( doc, node, attval, MISSING_ATTR_VALUE);
return;
}
/* Copy the attribute value so we can split it */
if (attval->value) {
tmbstr t, tPtr ;
uint prefixCount = 0;
/* isPrefix toggles - start at 1 and change to 0 as we
* iterate over the components of the value */
uint isPrefix = 1;
/* Copy it over */
uint len = TY_(tmbstrlen)(attval->value);
Fix a tiny buffer overflow.
2015-11-23 20:28:00 +00:00
tmbstr s = (tmbstr) TidyDocAlloc( doc, len + 1 );
Initial cut at RDFa support (again) New branch that implements support for RDFa attributes. Should be cleaner than my first attempt in PR #299 - also references issue #209
2015-11-16 17:29:23 +00:00
s[0] = '\0';
TY_(tmbstrcpy)( s, attval->value );
/* iterate over value */
tPtr = s;
while ( ( t = strtok(tPtr, " ") ) != NULL ) {
tPtr = NULL;
if (isPrefix) {
/* this piece should be a prefix */
/* prefix rules are that it can have any
* character except a colon - that one must be
* at the end */
tmbstr i = strchr(t, ':') ;
if (i == NULL) {
/* no colon - bad! */
TY_(ReportAttrError)( doc, node, attval, BAD_ATTRIBUTE_VALUE);
} else if (i != ( t + TY_(tmbstrlen)(t) - 1) ) {
/* not at the end - also bad */
TY_(ReportAttrError)( doc, node, attval, BAD_ATTRIBUTE_VALUE);
}
} else {
/* this piece should be a URL */
prefixCount ++;
}
isPrefix = !isPrefix;
}
TidyDocFree( doc, s ) ;
}
}
/* CheckRDFaTerm - are terms valid
*
*/
void CheckRDFaTerm ( TidyDocImpl* doc, Node *node, AttVal *attval)
{
if (!AttrHasValue(attval))
{
TY_(ReportAttrError)( doc, node, attval, MISSING_ATTR_VALUE);
return;
}
}
/* CheckRDFaSafeCURIE - is a CURIE legal
*
*/
void CheckRDFaSafeCURIE ( TidyDocImpl* doc, Node *node, AttVal *attval)
{
if (!AttrHasValue(attval))
{
TY_(ReportAttrError)( doc, node, attval, MISSING_ATTR_VALUE);
return;
}
}
new
2011-11-17 02:44:16 +00:00
/*
* local variables:
* mode: c
* indent-tabs-mode: nil
* c-basic-offset: 4
* eval: (c-set-offset 'substatement-open 0)
* end:
*/
Reference in a new issue Copy permalink