tris/recoll
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

html: small additional cleanup after previous <body> processing modification

Browse Source
This commit is contained in:
Jean-Francois Dockes 2012-05-16 10:13:53 +02:00
parent e6191b51a8
commit 0333d83d2e
1 changed files with 6 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

14
src/internfile/myhtmlparse.cpp
View File

@ -311,6 +311,7 @@ MyHtmlParser::opening_tag(const string &tag)
// even text before the body is displayed by Opera and // even text before the body is displayed by Opera and
// Firefox. We used to reset the dump each time we saw a // Firefox. We used to reset the dump each time we saw a
// body tag, but I can't see any reason to do so. // body tag, but I can't see any reason to do so.
if (tag == "blockquote" || tag == "br") { if (tag == "blockquote" || tag == "br") {
dump += '\n'; dump += '\n';
pending_space = true; pending_space = true;
@ -471,14 +472,11 @@ MyHtmlParser::closing_tag(const string &tag)
if (tag == "address") pending_space = true; if (tag == "address") pending_space = true;
break; break;
case 'b': case 'b':
if (tag == "body") { // body: We used to signal and end of doc here by returning
LOGDEB1(("Myhtmlparse: body close tag found\n")); // false but the browsers just ignore body and html
// We used to signal and end of doc here by returning // closing tags if there is further text, so it seems right
// false but the browsers just ignore body and html // to do the same
// closing tags if there is further text, so it seems right
// to do the same
break;
}
if (tag == "blockquote" || tag == "br") pending_space = true; if (tag == "blockquote" || tag == "br") pending_space = true;
break; break;
case 'c': case 'c':