Mercurial > emacs
comparison src/coding.c @ 21744:64c815fe1bdc
(shrink_decoding_region): Do not consider LF as ascii
if preceded by CR, since that confuses eol decoding.
(code_convert_region): When conversion fails with
CODING_FINISH_INSUFFICIENT_SRC, was overwriting src with garbage
from dst instead of copying from src to dst.
| author | Richard M. Stallman <rms@gnu.org> |
|---|---|
| date | Fri, 24 Apr 1998 01:05:25 +0000 |
| parents | 4097e558fa19 |
| children | 376ce06ab12d |
comparison
equal
deleted
inserted
replaced
| 21743:8def83521606 | 21744:64c815fe1bdc |
|---|---|
| 3780 case coding_type_raw_text: | 3780 case coding_type_raw_text: |
| 3781 if (eol_conversion) | 3781 if (eol_conversion) |
| 3782 { | 3782 { |
| 3783 if (coding->heading_ascii < 0) | 3783 if (coding->heading_ascii < 0) |
| 3784 while (begp < endp && *begp != '\r' && *begp < 0x80) begp++; | 3784 while (begp < endp && *begp != '\r' && *begp < 0x80) begp++; |
| 3785 while (begp < endp && *(endp - 1) != '\r' && *(endp - 1) < 0x80) | 3785 while (begp < endp && endp[-1] != '\r' && endp[-1] < 0x80) |
| 3786 endp--; | 3786 endp--; |
| 3787 /* Do not consider LF as ascii if preceded by CR, since that | |
| 3788 confuses eol decoding. */ | |
| 3789 if (begp < endp && endp < endp_orig && endp[-1] == '\r' && endp[0] == '\n') | |
| 3790 endp++; | |
| 3787 } | 3791 } |
| 3788 else | 3792 else |
| 3789 begp = endp; | 3793 begp = endp; |
| 3790 break; | 3794 break; |
| 3791 | 3795 |
| 3803 second byte of SJIS or BIG5 code. */ | 3807 second byte of SJIS or BIG5 code. */ |
| 3804 if (eol_conversion) | 3808 if (eol_conversion) |
| 3805 while (begp < endp && endp[-1] < 0x80 && endp[-1] != '\r') endp--; | 3809 while (begp < endp && endp[-1] < 0x80 && endp[-1] != '\r') endp--; |
| 3806 else | 3810 else |
| 3807 while (begp < endp && endp[-1] < 0x80) endp--; | 3811 while (begp < endp && endp[-1] < 0x80) endp--; |
| 3812 /* Do not consider LF as ascii if preceded by CR, since that | |
| 3813 confuses eol decoding. */ | |
| 3814 if (begp < endp && endp < endp_orig && endp[-1] == '\r' && endp[0] == '\n') | |
| 3815 endp++; | |
| 3808 if (begp < endp && endp < endp_orig && endp[-1] >= 0x80) | 3816 if (begp < endp && endp < endp_orig && endp[-1] >= 0x80) |
| 3809 endp++; | 3817 endp++; |
| 3810 break; | 3818 break; |
| 3811 | 3819 |
| 3812 default: /* i.e. case coding_type_iso2022: */ | 3820 default: /* i.e. case coding_type_iso2022: */ |
| 3827 /* We can skip all ASCII characters at the tail. */ | 3835 /* We can skip all ASCII characters at the tail. */ |
| 3828 if (eol_conversion) | 3836 if (eol_conversion) |
| 3829 while (begp < endp && (c = endp[-1]) < 0x80 && c != '\r') endp--; | 3837 while (begp < endp && (c = endp[-1]) < 0x80 && c != '\r') endp--; |
| 3830 else | 3838 else |
| 3831 while (begp < endp && endp[-1] < 0x80) endp--; | 3839 while (begp < endp && endp[-1] < 0x80) endp--; |
| 3840 /* Do not consider LF as ascii if preceded by CR, since that | |
| 3841 confuses eol decoding. */ | |
| 3842 if (begp < endp && endp < endp_orig && endp[-1] == '\r' && endp[0] == '\n') | |
| 3843 endp++; | |
| 3832 break; | 3844 break; |
| 3833 | 3845 |
| 3834 case CODING_CATEGORY_IDX_ISO_7: | 3846 case CODING_CATEGORY_IDX_ISO_7: |
| 3835 case CODING_CATEGORY_IDX_ISO_7_TIGHT: | 3847 case CODING_CATEGORY_IDX_ISO_7_TIGHT: |
| 3836 /* We can skip all charactes at the tail except for ESC and | 3848 /* We can skip all charactes at the tail except for ESC and |
| 3841 endp--; | 3853 endp--; |
| 3842 else | 3854 else |
| 3843 while (begp < endp | 3855 while (begp < endp |
| 3844 && (c = endp[-1]) < 0x80 && c != ISO_CODE_ESC) | 3856 && (c = endp[-1]) < 0x80 && c != ISO_CODE_ESC) |
| 3845 endp--; | 3857 endp--; |
| 3858 /* Do not consider LF as ascii if preceded by CR, since that | |
| 3859 confuses eol decoding. */ | |
| 3860 if (begp < endp && endp < endp_orig && endp[-1] == '\r' && endp[0] == '\n') | |
| 3861 endp++; | |
| 3846 if (begp < endp && endp[-1] == ISO_CODE_ESC) | 3862 if (begp < endp && endp[-1] == ISO_CODE_ESC) |
| 3847 { | 3863 { |
| 3848 if (endp + 1 < endp_orig && end[0] == '(' && end[1] == 'B') | 3864 if (endp + 1 < endp_orig && end[0] == '(' && end[1] == 'B') |
| 3849 /* This is an ASCII designation sequence. We can | 3865 /* This is an ASCII designation sequence. We can |
| 3850 surely skip the tail. */ | 3866 surely skip the tail. */ |
| 4220 /* The source text ends in invalid codes. Let's just | 4236 /* The source text ends in invalid codes. Let's just |
| 4221 make them valid buffer contents, and finish conversion. */ | 4237 make them valid buffer contents, and finish conversion. */ |
| 4222 inserted += len_byte; | 4238 inserted += len_byte; |
| 4223 inserted_byte += len_byte; | 4239 inserted_byte += len_byte; |
| 4224 while (len_byte--) | 4240 while (len_byte--) |
| 4225 *src++ = *dst++; | 4241 *dst++ = *src++; |
| 4226 fake_multibyte = 1; | 4242 fake_multibyte = 1; |
| 4227 break; | 4243 break; |
| 4228 } | 4244 } |
| 4229 if (first) | 4245 if (first) |
| 4230 { | 4246 { |