X-Git-Url: https://git.octo.it/?a=blobdiff_plain;f=count-delta.c;h=3ee3a0ccf1b92f3f8838ca2e728bde9f60d1b8e8;hb=6d5e6fff52968443018af56e2966cf5866130a4e;hp=e10c832a758f49e9117a3456686d54f8b94ce10b;hpb=8b7d510fb182ca40bd0f905bb2f5543b0d9549de;p=git.git diff --git a/count-delta.c b/count-delta.c index e10c832a..3ee3a0cc 100644 --- a/count-delta.c +++ b/count-delta.c @@ -3,58 +3,103 @@ * The delta-parsing part is almost straight copy of patch-delta.c * which is (C) 2005 Nicolas Pitre . */ +#include "cache.h" +#include "delta.h" +#include "count-delta.h" #include #include #include -#include "count-delta.h" -static unsigned long get_hdr_size(const unsigned char **datap) +struct span { + struct span *next; + unsigned long ofs; + unsigned long end; +}; + +static void touch_range(struct span **span, + unsigned long ofs, unsigned long end) { - const unsigned char *data = *datap; - unsigned long size; - unsigned char cmd; - int i; - size = i = 0; - cmd = *data++; - while (cmd) { - if (cmd & 1) - size |= *data++ << i; - i += 8; - cmd >>= 1; + struct span *e = *span; + struct span *p = NULL; + + while (e && e->ofs <= ofs) { + again: + if (ofs < e->end) { + while (e->end < end) { + if (e->next && e->next->ofs <= end) { + e->end = e->next->ofs; + e = e->next; + } + else { + e->end = end; + return; + } + } + return; + } + p = e; + e = e->next; + } + if (e && e->ofs <= end) { + e->ofs = ofs; + goto again; + } + else { + e = xmalloc(sizeof(*e)); + e->ofs = ofs; + e->end = end; + if (p) { + e->next = p->next; + p->next = e; + } + else { + e->next = *span; + *span = e; + } } - *datap = data; - return size; +} + +static unsigned long count_range(struct span *s) +{ + struct span *t; + unsigned long sz = 0; + while (s) { + t = s; + sz += s->end - s->ofs; + s = s->next; + free(t); + } + return sz; } /* * NOTE. We do not _interpret_ delta fully. As an approximation, we * just count the number of bytes that are copied from the source, and - * the number of literal data bytes that are inserted. Number of - * bytes that are _not_ copied from the source is deletion, and number - * of inserted literal bytes are addition, so sum of them is what we - * return. xdelta can express an edit that copies data inside of the - * destination which originally came from the source. We do not count - * that in the following routine, so we are undercounting the source - * material that remains in the final output that way. + * the number of literal data bytes that are inserted. + * + * Number of bytes that are _not_ copied from the source is deletion, + * and number of inserted literal bytes are addition, so sum of them + * is the extent of damage. */ -unsigned long count_delta(void *delta_buf, unsigned long delta_size) +int count_delta(void *delta_buf, unsigned long delta_size, + unsigned long *src_copied, unsigned long *literal_added) { - unsigned long copied_from_source, added_literal; + unsigned long added_literal; const unsigned char *data, *top; unsigned char cmd; unsigned long src_size, dst_size, out; + struct span *span = NULL; - /* the smallest delta size possible is 6 bytes */ - if (delta_size < 6) - return UINT_MAX; + if (delta_size < DELTA_SIZE_MIN) + return -1; data = delta_buf; top = delta_buf + delta_size; - src_size = get_hdr_size(&data); - dst_size = get_hdr_size(&data); + src_size = get_delta_hdr_size(&data); + dst_size = get_delta_hdr_size(&data); - added_literal = copied_from_source = out = 0; + added_literal = out = 0; while (data < top) { cmd = *data++; if (cmd & 0x80) { @@ -65,13 +110,10 @@ unsigned long count_delta(void *delta_buf, unsigned long delta_size) if (cmd & 0x08) cp_off |= (*data++ << 24); if (cmd & 0x10) cp_size = *data++; if (cmd & 0x20) cp_size |= (*data++ << 8); + if (cmd & 0x40) cp_size |= (*data++ << 16); if (cp_size == 0) cp_size = 0x10000; - if (cmd & 0x40) - /* copy from dst */ - ; - else - copied_from_source += cp_size; + touch_range(&span, cp_off, cp_off+cp_size); out += cp_size; } else { /* write literal into dst */ @@ -81,15 +123,15 @@ unsigned long count_delta(void *delta_buf, unsigned long delta_size) } } + *src_copied = count_range(span); + /* sanity check */ if (data != top || out != dst_size) - return UINT_MAX; + return -1; /* delete size is what was _not_ copied from source. * edit size is that and literal additions. */ - if (src_size + added_literal < copied_from_source) - /* we ended up overcounting and underflowed */ - return 0; - return (src_size - copied_from_source) + added_literal; + *literal_added = added_literal; + return 0; }