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

small amd64 fixes: 64 bits size_type, signed chars

Browse Source
This commit is contained in:
dockes 2009-12-13 16:13:59 +00:00
parent 658a21d5ad
commit bf3ac8e053
4 changed files with 8 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
src/common/textsplit.cpp
View File

@ -50,9 +50,10 @@ using namespace std;
// The array is actually a remnant of the original version which did no utf8. // The array is actually a remnant of the original version which did no utf8.
// Only the lower 127 slots are now used, but keep it at 256 // Only the lower 127 slots are now used, but keep it at 256
// because it makes some tests in the code simpler. // because it makes some tests in the code simpler.
const unsigned int charclasses_size = 256;
enum CharClass {LETTER=256, SPACE=257, DIGIT=258, WILD=259, enum CharClass {LETTER=256, SPACE=257, DIGIT=258, WILD=259,
A_ULETTER=260, A_LLETTER=261}; A_ULETTER=260, A_LLETTER=261};
static int charclasses[256]; static int charclasses[charclasses_size];
// Real UTF-8 characters are handled with sets holding all characters // Real UTF-8 characters are handled with sets holding all characters
// with interesting properties. This is far from full-blown management // with interesting properties. This is far from full-blown management
@ -454,7 +455,7 @@ bool TextSplit::text_to_words(const string &in)
// This emits "camel" when hitting the 'C' of camelCase // This emits "camel" when hitting the 'C' of camelCase
case A_ULETTER: case A_ULETTER:
if (m_span.length() && if (m_span.length() &&
charclasses[(unsigned int)m_span[m_span.length() - 1]] == charclasses[(unsigned char)m_span[m_span.length() - 1]] ==
A_LLETTER) { A_LLETTER) {
if (m_wordLen) { if (m_wordLen) {
if (!doemit(false, it.getBpos())) if (!doemit(false, it.getBpos()))
@ -471,7 +472,7 @@ bool TextSplit::text_to_words(const string &in)
// Emit the uppercase word before proceeding // Emit the uppercase word before proceeding
case A_LLETTER: case A_LLETTER:
if (m_span.length() && if (m_span.length() &&
charclasses[(unsigned int)m_span[m_span.length() - 1]] == charclasses[(unsigned char)m_span[m_span.length() - 1]] ==
A_ULETTER && m_wordLen > 1) { A_ULETTER && m_wordLen > 1) {
// Multiple upper-case letters. Single letter word // Multiple upper-case letters. Single letter word
// or acronym which we want to emit now // or acronym which we want to emit now
@ -611,7 +612,7 @@ bool TextSplit::hasVisibleWhite(const string &in)
setcharclasses(); setcharclasses();
Utf8Iter it(in); Utf8Iter it(in);
for (; !it.eof(); it++) { for (; !it.eof(); it++) {
unsigned int c = *it; unsigned int c = (unsigned char)*it;
LOGDEB3(("TextSplit::hasVisibleWhite: testing 0x%04x\n", c)); LOGDEB3(("TextSplit::hasVisibleWhite: testing 0x%04x\n", c));
if (c == (unsigned int)-1) { if (c == (unsigned int)-1) {
LOGERR(("hasVisibleWhite: error while scanning UTF-8 string\n")); LOGERR(("hasVisibleWhite: error while scanning UTF-8 string\n"));

2
src/internfile/mh_html.cpp
View File

@ -131,7 +131,7 @@ bool MimeHandlerHtml::next_document()
// because the file was transcoded. It seems that just // because the file was transcoded. It seems that just
// inserting one is enough (only the 1st one seems to // inserting one is enough (only the 1st one seems to
// be used by browsers/qtextedit). // be used by browsers/qtextedit).
unsigned int idx = m_html.find("<head>"); string::size_type idx = m_html.find("<head>");
if (idx == string::npos) if (idx == string::npos)
idx = m_html.find("<HEAD>"); idx = m_html.find("<HEAD>");
if (idx != string::npos) if (idx != string::npos)

2
src/query/plaintorich.cpp
View File

@ -378,7 +378,7 @@ bool PlainToRich::plaintorich(const string& in,
// My tag state // My tag state
int inrcltag = 0; int inrcltag = 0;
unsigned int headend = 0; string::size_type headend = 0;
if (m_inputhtml) { if (m_inputhtml) {
headend = in.find("</head>"); headend = in.find("</head>");
if (headend == string::npos) if (headend == string::npos)

2
src/utils/base64.cpp
View File

@ -125,7 +125,7 @@ bool base64_decode(const string& in, string& out)
for (; ii < in.length(); ch = in[ii++]) for (; ii < in.length(); ch = in[ii++])
if (!isspace((unsigned char)ch)) { if (!isspace((unsigned char)ch)) {
DPRINT((stderr, "base64_dec: non-white at eod: 0x%x\n", DPRINT((stderr, "base64_dec: non-white at eod: 0x%x\n",
(unsigned int)ch)); (unsigned int)((unsigned char)ch)));
// Well, there are bad encoders out there. Let it pass // Well, there are bad encoders out there. Let it pass
//return false; //return false;
} }