Parse conflicts/provides/replaces at database load time

We did this with depends way back in commit c244cfecf654d3 in 2007. We
can do it with these fields as well.

Of note is the inclusion of provides even though only '=' is supported-
we'll parse other things, but no guarantees are given as to behavior,
which is more or less similar to before since we only looked for the
equals sign.

Also of note is the non-inclusion of optdepends; this will likely be
resolved down the road.

The biggest benefactors of this change will be the resolving code that
formerly had to parse and reparse several of these fields; it only
happens once now at load time. This does lead to the disadvantage that
we will now always be parsing this information up front even if we never
need it in the split form, but as these are uncommon fields and our
parser is quite efficient it shouldn't be a big concern.

Signed-off-by: Dan McGee <dan@archlinux.org>

9 files changed, 83 insertions, 71 deletions

diff --git a/lib/libalpm/alpm.h b/lib/libalpm/alpm.h
index 3febd0ec..fc8f0bcd 100644
--- a/lib/libalpm/alpm.h
+++ b/lib/libalpm/alpm.h
@@ -155,7 +155,7 @@ typedef struct _alpm_conflict_t {
 	unsigned long package2_hash;
 	char *package1;
 	char *package2;
-	char *reason;
+	alpm_depend_t *reason;
 } alpm_conflict_t;
 
 /** File conflict */
@@ -670,15 +670,15 @@ alpm_list_t *alpm_pkg_get_depends(alpm_pkg_t *pkg);
  */
 alpm_list_t *alpm_pkg_get_optdepends(alpm_pkg_t *pkg);
 
-/** Returns the list of package names conflicting with pkg.
+/** Returns the list of packages conflicting with pkg.
  * @param pkg a pointer to package
- * @return a reference to an internal list of strings.
+ * @return a reference to an internal list of alpm_depend_t structures.
  */
 alpm_list_t *alpm_pkg_get_conflicts(alpm_pkg_t *pkg);
 
-/** Returns the list of package names provided by pkg.
+/** Returns the list of packages provided by pkg.
  * @param pkg a pointer to package
- * @return a reference to an internal list of strings.
+ * @return a reference to an internal list of alpm_depend_t structures.
  */
 alpm_list_t *alpm_pkg_get_provides(alpm_pkg_t *pkg);
 
@@ -690,7 +690,7 @@ alpm_list_t *alpm_pkg_get_deltas(alpm_pkg_t *pkg);
 
 /** Returns the list of packages to be replaced by pkg.
  * @param pkg a pointer to package
- * @return a reference to an internal list of strings.
+ * @return a reference to an internal list of alpm_depend_t structures.
  */
 alpm_list_t *alpm_pkg_get_replaces(alpm_pkg_t *pkg);
 
diff --git a/lib/libalpm/be_local.c b/lib/libalpm/be_local.c
index fdb4ceaf..a874504e 100644
--- a/lib/libalpm/be_local.c
+++ b/lib/libalpm/be_local.c
@@ -509,6 +509,12 @@ static char *get_pkgpath(alpm_db_t *db, alpm_pkg_t *info)
 	f = alpm_list_add(f, linedup); \
 } while(1) /* note the while(1) and not (0) */
 
+#define READ_AND_SPLITDEP(f) do { \
+	if(fgets(line, sizeof(line), fp) == NULL && !feof(fp)) goto error; \
+	if(_alpm_strip_newline(line) == 0) break; \
+	f = alpm_list_add(f, _alpm_splitdep(line)); \
+} while(1) /* note the while(1) and not (0) */
+
 static int local_db_read(alpm_pkg_t *info, alpm_dbinfrq_t inforeq)
 {
 	FILE *fp = NULL;
@@ -601,20 +607,15 @@ static int local_db_read(alpm_pkg_t *info, alpm_dbinfrq_t inforeq)
 				/* also store this value to isize */
 				info->isize = info->size;
 			} else if(strcmp(line, "%REPLACES%") == 0) {
-				READ_AND_STORE_ALL(info->replaces);
+				READ_AND_SPLITDEP(info->replaces);
 			} else if(strcmp(line, "%DEPENDS%") == 0) {
-				/* Different than the rest because of the _alpm_splitdep call. */
-				while(1) {
-					READ_NEXT();
-					if(strlen(line) == 0) break;
-					info->depends = alpm_list_add(info->depends, _alpm_splitdep(line));
-				}
+				READ_AND_SPLITDEP(info->depends);
 			} else if(strcmp(line, "%OPTDEPENDS%") == 0) {
 				READ_AND_STORE_ALL(info->optdepends);
 			} else if(strcmp(line, "%CONFLICTS%") == 0) {
-				READ_AND_STORE_ALL(info->conflicts);
+				READ_AND_SPLITDEP(info->conflicts);
 			} else if(strcmp(line, "%PROVIDES%") == 0) {
-				READ_AND_STORE_ALL(info->provides);
+				READ_AND_SPLITDEP(info->provides);
 			}
 		}
 		fclose(fp);
@@ -771,7 +772,9 @@ int _alpm_local_db_write(alpm_db_t *db, alpm_pkg_t *info, alpm_dbinfrq_t inforeq
 		if(info->replaces) {
 			fputs("%REPLACES%\n", fp);
 			for(lp = info->replaces; lp; lp = lp->next) {
-				fprintf(fp, "%s\n", (char *)lp->data);
+				char *depstring = alpm_dep_compute_string(lp->data);
+				fprintf(fp, "%s\n", depstring);
+				free(depstring);
 			}
 			fprintf(fp, "\n");
 		}
@@ -830,14 +833,18 @@ int _alpm_local_db_write(alpm_db_t *db, alpm_pkg_t *info, alpm_dbinfrq_t inforeq
 		if(info->conflicts) {
 			fputs("%CONFLICTS%\n", fp);
 			for(lp = info->conflicts; lp; lp = lp->next) {
-				fprintf(fp, "%s\n", (char *)lp->data);
+				char *depstring = alpm_dep_compute_string(lp->data);
+				fprintf(fp, "%s\n", depstring);
+				free(depstring);
 			}
 			fprintf(fp, "\n");
 		}
 		if(info->provides) {
 			fputs("%PROVIDES%\n", fp);
 			for(lp = info->provides; lp; lp = lp->next) {
-				fprintf(fp, "%s\n", (char *)lp->data);
+				char *depstring = alpm_dep_compute_string(lp->data);
+				fprintf(fp, "%s\n", depstring);
+				free(depstring);
 			}
 			fprintf(fp, "\n");
 		}
diff --git a/lib/libalpm/be_package.c b/lib/libalpm/be_package.c
index 9762cbca..3b5b0d0c 100644
--- a/lib/libalpm/be_package.c
+++ b/lib/libalpm/be_package.c
@@ -194,11 +194,14 @@ static int parse_descfile(alpm_handle_t *handle, struct archive *a, alpm_pkg_t *
 			} else if(strcmp(key, "optdepend") == 0) {
 				newpkg->optdepends = alpm_list_add(newpkg->optdepends, strdup(ptr));
 			} else if(strcmp(key, "conflict") == 0) {
-				newpkg->conflicts = alpm_list_add(newpkg->conflicts, strdup(ptr));
+				alpm_depend_t *conflict = _alpm_splitdep(ptr);
+				newpkg->conflicts = alpm_list_add(newpkg->conflicts, conflict);
 			} else if(strcmp(key, "replaces") == 0) {
-				newpkg->replaces = alpm_list_add(newpkg->replaces, strdup(ptr));
+				alpm_depend_t *replace = _alpm_splitdep(ptr);
+				newpkg->replaces = alpm_list_add(newpkg->replaces, replace);
 			} else if(strcmp(key, "provides") == 0) {
-				newpkg->provides = alpm_list_add(newpkg->provides, strdup(ptr));
+				alpm_depend_t *provide = _alpm_splitdep(ptr);
+				newpkg->provides = alpm_list_add(newpkg->provides, provide);
 			} else if(strcmp(key, "backup") == 0) {
 				alpm_backup_t *backup;
 				CALLOC(backup, 1, sizeof(alpm_backup_t), return -1);
diff --git a/lib/libalpm/be_sync.c b/lib/libalpm/be_sync.c
index 069e39dd..66808a7f 100644
--- a/lib/libalpm/be_sync.c
+++ b/lib/libalpm/be_sync.c
@@ -460,6 +460,12 @@ static int sync_db_populate(alpm_db_t *db)
 	f = alpm_list_add(f, linedup); \
 } while(1) /* note the while(1) and not (0) */
 
+#define READ_AND_SPLITDEP(f) do { \
+	if(_alpm_archive_fgets(archive, &buf) != ARCHIVE_OK) goto error; \
+	if(_alpm_strip_newline(buf.line) == 0) break; \
+	f = alpm_list_add(f, _alpm_splitdep(line)); \
+} while(1) /* note the while(1) and not (0) */
+
 static int sync_db_read(alpm_db_t *db, struct archive *archive,
 		struct archive_entry *entry, alpm_pkg_t **likely_pkg)
 {
@@ -547,20 +553,15 @@ static int sync_db_read(alpm_db_t *db, struct archive *archive,
 			} else if(strcmp(line, "%PGPSIG%") == 0) {
 				READ_AND_STORE(pkg->base64_sig);
 			} else if(strcmp(line, "%REPLACES%") == 0) {
-				READ_AND_STORE_ALL(pkg->replaces);
+				READ_AND_SPLITDEP(pkg->replaces);
 			} else if(strcmp(line, "%DEPENDS%") == 0) {
-				/* Different than the rest because of the _alpm_splitdep call. */
-				while(1) {
-					READ_NEXT();
-					if(strlen(line) == 0) break;
-					pkg->depends = alpm_list_add(pkg->depends, _alpm_splitdep(line));
-				}
+				READ_AND_SPLITDEP(pkg->depends);
 			} else if(strcmp(line, "%OPTDEPENDS%") == 0) {
 				READ_AND_STORE_ALL(pkg->optdepends);
 			} else if(strcmp(line, "%CONFLICTS%") == 0) {
-				READ_AND_STORE_ALL(pkg->conflicts);
+				READ_AND_SPLITDEP(pkg->conflicts);
 			} else if(strcmp(line, "%PROVIDES%") == 0) {
-				READ_AND_STORE_ALL(pkg->provides);
+				READ_AND_SPLITDEP(pkg->provides);
 			} else if(strcmp(line, "%DELTAS%") == 0) {
 				/* Different than the rest because of the _alpm_delta_parse call. */
 				while(1) {
diff --git a/lib/libalpm/conflict.c b/lib/libalpm/conflict.c
index 91e6b677..fe182094 100644
--- a/lib/libalpm/conflict.c
+++ b/lib/libalpm/conflict.c
@@ -42,7 +42,7 @@
 #include "deps.h"
 
 static alpm_conflict_t *conflict_new(alpm_pkg_t *pkg1, alpm_pkg_t *pkg2,
-		const char *reason)
+		alpm_depend_t *reason)
 {
 	alpm_conflict_t *conflict;
 
@@ -52,7 +52,7 @@ static alpm_conflict_t *conflict_new(alpm_pkg_t *pkg1, alpm_pkg_t *pkg2,
 	conflict->package2_hash = pkg2->name_hash;
 	STRDUP(conflict->package1, pkg1->name, return NULL);
 	STRDUP(conflict->package2, pkg2->name, return NULL);
-	STRDUP(conflict->reason, reason, return NULL);
+	conflict->reason = reason;
 
 	return conflict;
 }
@@ -61,7 +61,6 @@ void _alpm_conflict_free(alpm_conflict_t *conflict)
 {
 	FREE(conflict->package2);
 	FREE(conflict->package1);
-	FREE(conflict->reason);
 	FREE(conflict);
 }
 
@@ -74,7 +73,7 @@ alpm_conflict_t *_alpm_conflict_dup(const alpm_conflict_t *conflict)
 	newconflict->package2_hash = conflict->package2_hash;
 	STRDUP(newconflict->package1, conflict->package1, return NULL);
 	STRDUP(newconflict->package2, conflict->package2, return NULL);
-	STRDUP(newconflict->reason, conflict->reason, return NULL);
+	newconflict->reason = conflict->reason;
 
 	return newconflict;
 }
@@ -103,16 +102,18 @@ static int conflict_isin(alpm_conflict_t *needle, alpm_list_t *haystack)
  * @param reason reason for this conflict
  */
 static int add_conflict(alpm_handle_t *handle, alpm_list_t **baddeps,
-		alpm_pkg_t *pkg1, alpm_pkg_t *pkg2, const char *reason)
+		alpm_pkg_t *pkg1, alpm_pkg_t *pkg2, alpm_depend_t *reason)
 {
 	alpm_conflict_t *conflict = conflict_new(pkg1, pkg2, reason);
 	if(!conflict) {
 		return -1;
 	}
-	_alpm_log(handle, ALPM_LOG_DEBUG, "package %s conflicts with %s (by %s)\n",
-			pkg1->name, pkg2->name, reason);
 	if(!conflict_isin(conflict, *baddeps)) {
+		char *conflict_str = alpm_dep_compute_string(reason);
 		*baddeps = alpm_list_add(*baddeps, conflict);
+		_alpm_log(handle, ALPM_LOG_DEBUG, "package %s conflicts with %s (by %s)\n",
+				pkg1->name, pkg2->name, conflict_str);
+		free(conflict_str);
 	} else {
 		_alpm_conflict_free(conflict);
 	}
@@ -144,9 +145,8 @@ static void check_conflict(alpm_handle_t *handle,
 		alpm_list_t *j;
 
 		for(j = alpm_pkg_get_conflicts(pkg1); j; j = j->next) {
-			const char *conflict = j->data;
+			alpm_depend_t *conflict = j->data;
 			alpm_list_t *k;
-			alpm_depend_t *parsed_conflict = _alpm_splitdep(conflict);
 
 			for(k = list2; k; k = k->next) {
 				alpm_pkg_t *pkg2 = k->data;
@@ -157,7 +157,7 @@ static void check_conflict(alpm_handle_t *handle,
 					continue;
 				}
 
-				if(_alpm_depcmp(pkg2, parsed_conflict)) {
+				if(_alpm_depcmp(pkg2, conflict)) {
 					if(order >= 0) {
 						add_conflict(handle, baddeps, pkg1, pkg2, conflict);
 					} else {
@@ -165,7 +165,6 @@ static void check_conflict(alpm_handle_t *handle,
 					}
 				}
 			}
-			_alpm_dep_free(parsed_conflict);
 		}
 	}
 }
diff --git a/lib/libalpm/db.c b/lib/libalpm/db.c
index 9e8e75bf..5765ab2c 100644
--- a/lib/libalpm/db.c
+++ b/lib/libalpm/db.c
@@ -437,7 +437,8 @@ alpm_list_t *_alpm_db_search(alpm_db_t *db, const alpm_list_t *needles)
 			if(!matched) {
 				/* check provides */
 				for(k = alpm_pkg_get_provides(pkg); k; k = k->next) {
-					if(regexec(&reg, k->data, 0, 0, 0) == 0) {
+					alpm_depend_t *provide = k->data;
+					if(regexec(&reg, provide->name, 0, 0, 0) == 0) {
 						matched = k->data;
 						break;
 					}
diff --git a/lib/libalpm/deps.c b/lib/libalpm/deps.c
index 992ebe23..6869087c 100644
--- a/lib/libalpm/deps.c
+++ b/lib/libalpm/deps.c
@@ -384,25 +384,19 @@ int _alpm_depcmp(alpm_pkg_t *pkg, alpm_depend_t *dep)
 		}
 	}
 
-	/* check provisions, format : "name=version" */
+	/* check provisions, name and version if available */
 	for(i = alpm_pkg_get_provides(pkg); i && !satisfy; i = i->next) {
-		const char *provision = i->data;
-		const char *provver = strchr(provision, '=');
-
-		if(provver == NULL) { /* no provision version */
-			satisfy = (dep->mod == ALPM_DEP_MOD_ANY
-					&& strcmp(provision, dep->name) == 0);
-		} else {
-			/* This is a bit tricker than the old code for performance reasons. To
-			 * prevent the need to copy and duplicate strings, strncmp only the name
-			 * portion if they are the same length, since there is a version and
-			 * operator in play here. Cast is to silence sign conversion warning;
-			 * we know provver >= provision if we are here. */
-			size_t namelen = (size_t)(provver - provision);
-			provver += 1;
-			satisfy = (strlen(dep->name) == namelen
-					&& strncmp(provision, dep->name, namelen) == 0
-					&& dep_vercmp(provver, dep->mod, dep->version));
+		alpm_depend_t *provision = i->data;
+
+		if(dep->mod == ALPM_DEP_MOD_ANY) {
+			/* any version will satisfy the requirement */
+			satisfy = (provision->name_hash == dep->name_hash
+					&& strcmp(provision->name, dep->name) == 0);
+		} else if (provision->mod == ALPM_DEP_MOD_EQ) {
+			/* provision specifies a version, so try it out */
+			satisfy = (provision->name_hash == dep->name_hash
+					&& strcmp(provision->name, dep->name) == 0
+					&& dep_vercmp(provision->version, dep->mod, dep->version));
 		}
 	}
 
diff --git a/lib/libalpm/package.c b/lib/libalpm/package.c
index a8dc1440..19d2c844 100644
--- a/lib/libalpm/package.c
+++ b/lib/libalpm/package.c
@@ -496,7 +496,9 @@ alpm_pkg_t *_alpm_pkg_dup(alpm_pkg_t *pkg)
 	newpkg->reason = pkg->reason;
 
 	newpkg->licenses   = alpm_list_strdup(pkg->licenses);
-	newpkg->replaces   = alpm_list_strdup(pkg->replaces);
+	for(i = pkg->replaces; i; i = alpm_list_next(i)) {
+		newpkg->replaces = alpm_list_add(newpkg->replaces, _alpm_dep_dup(i->data));
+	}
 	newpkg->groups     = alpm_list_strdup(pkg->groups);
 	if(pkg->files.count) {
 		size_t filenum;
@@ -517,8 +519,12 @@ alpm_pkg_t *_alpm_pkg_dup(alpm_pkg_t *pkg)
 		newpkg->depends = alpm_list_add(newpkg->depends, _alpm_dep_dup(i->data));
 	}
 	newpkg->optdepends = alpm_list_strdup(pkg->optdepends);
-	newpkg->conflicts  = alpm_list_strdup(pkg->conflicts);
-	newpkg->provides   = alpm_list_strdup(pkg->provides);
+	for(i = pkg->conflicts; i; i = alpm_list_next(i)) {
+		newpkg->conflicts = alpm_list_add(newpkg->conflicts, _alpm_dep_dup(i->data));
+	}
+	for(i = pkg->provides; i; i = alpm_list_next(i)) {
+		newpkg->provides = alpm_list_add(newpkg->provides, _alpm_dep_dup(i->data));
+	}
 	for(i = pkg->deltas; i; i = alpm_list_next(i)) {
 		newpkg->deltas = alpm_list_add(newpkg->deltas, _alpm_delta_dup(i->data));
 	}
@@ -557,8 +563,10 @@ void _alpm_pkg_free(alpm_pkg_t *pkg)
 	FREE(pkg->sha256sum);
 	FREE(pkg->base64_sig);
 	FREE(pkg->arch);
+
 	FREELIST(pkg->licenses);
-	FREELIST(pkg->replaces);
+	alpm_list_free_inner(pkg->replaces, (alpm_list_fn_free)_alpm_dep_free);
+	alpm_list_free(pkg->replaces);
 	FREELIST(pkg->groups);
 	if(pkg->files.count) {
 		size_t i;
@@ -572,8 +580,10 @@ void _alpm_pkg_free(alpm_pkg_t *pkg)
 	alpm_list_free_inner(pkg->depends, (alpm_list_fn_free)_alpm_dep_free);
 	alpm_list_free(pkg->depends);
 	FREELIST(pkg->optdepends);
-	FREELIST(pkg->conflicts);
-	FREELIST(pkg->provides);
+	alpm_list_free_inner(pkg->conflicts, (alpm_list_fn_free)_alpm_dep_free);
+	alpm_list_free(pkg->conflicts);
+	alpm_list_free_inner(pkg->provides, (alpm_list_fn_free)_alpm_dep_free);
+	alpm_list_free(pkg->provides);
 	alpm_list_free_inner(pkg->deltas, (alpm_list_fn_free)_alpm_delta_free);
 	alpm_list_free(pkg->deltas);
 	alpm_list_free(pkg->delta_path);
diff --git a/lib/libalpm/sync.c b/lib/libalpm/sync.c
index f16489d9..3e38707e 100644
--- a/lib/libalpm/sync.c
+++ b/lib/libalpm/sync.c
@@ -150,14 +150,11 @@ int SYMEXPORT alpm_sync_sysupgrade(alpm_handle_t *handle, int enable_downgrade)
 				for(k = _alpm_db_get_pkgcache(sdb); k; k = k->next) {
 					spkg = k->data;
 					for(l = alpm_pkg_get_replaces(spkg); l; l = l->next) {
-						const char *replace = l->data;
-						alpm_depend_t *parsed_replace = _alpm_splitdep(replace);
-						if(_alpm_depcmp(lpkg, parsed_replace)) {
+						alpm_depend_t *replace = l->data;
+						if(_alpm_depcmp(lpkg, replace)) {
 							found = 1;
-							_alpm_dep_free(parsed_replace);
 							break;
 						}
-						_alpm_dep_free(parsed_replace);
 					}
 					if(found) {
 						/* check IgnorePkg/IgnoreGroup */