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

Windows: text/plain: check if already utf-8 before trying transcode

Browse Source
This commit is contained in:
Jean-Francois Dockes 2020-08-15 11:16:10 +01:00
parent d9c1a9648c
commit cca69cbd31
1 changed files with 23 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

15
src/internfile/txtdcode.cpp
View File

@ -24,6 +24,10 @@
#include "smallut.h" #include "smallut.h"
#include "listmem.h" #include "listmem.h"
#ifdef _WIN32
#include "utf8iter.h"
#endif
using std::string; using std::string;
// Called after decoding from utf-8 failed. Handle the common case // Called after decoding from utf-8 failed. Handle the common case
@ -111,6 +115,17 @@ bool RecollFilter::txtdcode(const string& who)
int ecnt; int ecnt;
string otext; string otext;
#ifdef _WIN32
// Under Windows the environment charset will usually not be
// utf-8. We check if the text is actually utf-8. This is worth
// it, else the conversion from 8-bit is going to succeed if the
// text is already utf-8, and produce bogus data.
if (utf8check(itext, otext) >= 0) {
m_metaData[cstr_dj_keycharset] = cstr_utf8;
return true;
}
#endif
string bomfromcode = bomtocode(itext); string bomfromcode = bomtocode(itext);
if (!bomfromcode.empty()) { if (!bomfromcode.empty()) {
LOGDEB(who << "::txtdcode: " << " input charset changed from " << LOGDEB(who << "::txtdcode: " << " input charset changed from " <<