diff options
| author | Tatsuya Kinoshita <tats@debian.org> | 2021-01-02 00:20:47 +0000 |
|---|---|---|
| committer | Tatsuya Kinoshita <tats@debian.org> | 2021-01-02 00:20:47 +0000 |
| commit | a722a562112506c63c9b6519b8252d633534a35a (patch) | |
| tree | 78dc4b8bbefa910007339970eaf004f012d2082f /entity.c | |
| parent | Switch upstream to Debian's w3m (diff) | |
| parent | New upstream version 0.5.3+git20210102 (diff) | |
| download | w3m-a722a562112506c63c9b6519b8252d633534a35a.tar.gz w3m-a722a562112506c63c9b6519b8252d633534a35a.zip | |
Update upstream source from tag 'upstream/0.5.3+git20210102'
Update to upstream version '0.5.3+git20210102'
with Debian dir e89c066385156d7c9ed01b0c8a1a08e01019959d
Diffstat (limited to 'entity.c')
| -rw-r--r-- | entity.c | 16 |
1 files changed, 15 insertions, 1 deletions
@@ -44,6 +44,8 @@ conv_entity(unsigned int c) return " "; if (c == 0xa0) return NBSP; + if (c == 0xad) /* SOFT HYPHEN */ + return ""; if (c < 0x100) { /* Latin1 (ISO 8859-1) */ if (UseAltEntity) return alt_latin1[c - 0xa0]; @@ -56,11 +58,23 @@ conv_entity(unsigned int c) #ifdef USE_M17N #ifdef USE_UNICODE if (c <= WC_C_UCS4_END) { /* Unicode */ + char *chk; wc_uchar utf8[7]; wc_ucs_to_utf8(c, utf8); - return wc_conv((char *)utf8, WC_CES_UTF_8, InnerCharset)->ptr; + /* we eventually need to display it so check DisplayCharset */ + chk = wc_conv((char *)utf8, WC_CES_UTF_8, DisplayCharset ? DisplayCharset : WC_CES_US_ASCII)->ptr; + if (strcmp(chk, "?") != 0) + return wc_conv((char *)utf8, WC_CES_UTF_8, InnerCharset)->ptr; } #endif #endif + if (c == 0x201c || c == 0x201f || c == 0x201d || c == 0x2033) + return "\""; + if (c == 0x2018 || c == 0x201b || c == 0x2019 || c == 0x2032) + return "'"; + if (c >= 0x2010 && c < 0x2014) + return "-"; + if (c == 0x2014) + return "--"; return "?"; } |