github-actions[bot] commented on code in PR #67918:
URL: https://github.com/apache/doris/pull/67918#discussion_r4069689686
##########
fe/fe-core/src/main/java/org/apache/doris/analysis/invertedindex/AnalyzerIdentityBuilder.java:
##########
@@ -150,49 +221,458 @@ private static String
buildIdentityFromPolicyProperties(IndexPolicyTypeEnum type
* Resolve a component (tokenizer) to its identity.
*/
private static String resolveComponentIdentity(String name,
IndexPolicyTypeEnum expectedType) {
+ return resolveComponentIdentity(name, expectedType, null);
+ }
+
+ /**
+ * {@code foldBlockedBytes} is the case-folding context of a char filter:
null without a
+ * downstream fold, otherwise the bytes that filters between this one and
the fold rewrite.
+ */
+ private static String resolveComponentIdentity(
+ String name, IndexPolicyTypeEnum expectedType, boolean[]
foldBlockedBytes) {
if (Strings.isNullOrEmpty(name)) {
return "";
}
- // Check if it's a built-in component
- if (expectedType == IndexPolicyTypeEnum.TOKENIZER
- && IndexPolicy.BUILTIN_TOKENIZERS.contains(name)) {
- return name;
+ // Existing named policies take precedence over built-ins for upgrade
compatibility.
+ try {
+ Env env = Env.getCurrentEnv();
+ if (env != null && env.getIndexPolicyMgr() != null) {
+ IndexPolicy policy =
env.getIndexPolicyMgr().getPolicyByName(name);
+ if (policy != null && policy.getType() == expectedType) {
+ if (policy.isInvalid()) {
+ return "invalid-policy:" + policy.getId() + ":" +
policy.getName();
+ }
+ Map<String, String> props = policy.getProperties();
+ if (props != null && !props.isEmpty()) {
+ TreeMap<String, String> sortedProps = new
TreeMap<>(props);
+ String type = sortedProps.get(IndexPolicy.PROP_TYPE);
+ String normalizedType =
normalizeBuiltinComponentName(type, expectedType);
+ if (normalizedType != null) {
+ if ("empty".equals(normalizedType)) {
+ return "";
+ }
+ sortedProps.put(IndexPolicy.PROP_TYPE,
normalizedType);
+ canonicalizeEffectiveComponentProperties(
+ sortedProps, normalizedType, expectedType);
+ if (sortedProps.size() == 1) {
+ return normalizedType;
+ }
+ }
+ if (expectedType == IndexPolicyTypeEnum.TOKENIZER
+ &&
"ngram".equals(sortedProps.get(IndexPolicy.PROP_TYPE))) {
+ // This setting only limits policy creation; it
does not change emitted tokens.
+ sortedProps.remove(PROP_MAX_NGRAM_DIFF);
+ }
+ if (expectedType == IndexPolicyTypeEnum.CHAR_FILTER
+ &&
"char_replace".equals(sortedProps.get(IndexPolicy.PROP_TYPE))) {
+ String replacement =
sortedProps.getOrDefault("replacement", " ");
+ String pattern = canonicalizeCharReplacePattern(
+ sortedProps.get("pattern"), replacement,
foldBlockedBytes);
+ if (pattern.isEmpty()) {
+ return "";
+ }
+ sortedProps.put("pattern", pattern);
+ sortedProps.put("replacement", replacement);
+ }
+ if (normalizedType != null && sortedProps.size() == 1)
{
+ return normalizedType;
+ }
+ return sortedProps.toString();
+ }
+ }
+ }
+ } catch (RuntimeException e) {
+ // Fall through to built-in resolution or the original name.
+ }
+
+ String normalizedName = normalizeBuiltinComponentName(name,
expectedType);
+ return "empty".equals(normalizedName) ? "" : normalizedName == null ?
name : normalizedName;
+ }
+
+ private static void canonicalizeEffectiveComponentProperties(
+ TreeMap<String, String> properties, String type,
IndexPolicyTypeEnum expectedType) {
+ if ("pinyin".equals(type)) {
Review Comment:
[P1] Canonicalize the Pinyin tokenizer's `trim_whitespace` dependency on
`keep_original`. With the default `keep_original=false`, every emitted
tokenizer candidate is built from Chinese/pinyin or ASCII alphanumerics;
whitespace is skipped, and only the disabled original candidate could contain
leading/trailing whitespace. `{type=pinyin}` and
`{type=pinyin,trim_whitespace=false}` therefore emit the same terms, positions,
and offsets, yet receive different identities, so differently named analyzer
aliases can pass CREATE/ALTER duplicate-index checks. Drop this property for
TOKENIZER identities when the effective `keep_original` is false, retain it for
token filters/kept originals, and cover both DDL paths.
##########
be/src/storage/index/inverted/token_filter/pinyin_filter_factory.cpp:
##########
@@ -62,7 +62,10 @@ TokenFilterPtr PinyinFilterFactory::create(const
TokenStreamPtr& in) {
auto filter = std::make_shared<PinyinFilter>(in, config_);
filter->initialize();
+ if (!config_->ignorePinyinOffset) {
+ filter->set_source_byte_offsets_enabled(true);
Review Comment:
[P1] Include the Pinyin tokenizer in this provenance contract, and derive
candidate ranges from actual source positions. This enable call reaches
`PinyinTokenizer`, but it publishes no exact map or conservative span; `中 ->
zhong` therefore lets the following offset-aware Pinyin filter reconstruct five
byte positions beyond source `[0,3)`. Its current ranges are not safe to reuse:
`a-b` is compacted to `ab` but tagged `[0,2)` although `b` is at `[2,3)`, and a
preceding ICU filter can expand source U+FB01 `[0,3)` to `fi` while this
tokenizer leaves `[0,2)` because it never calls `correct_source_offset()`. This
is distinct from the earlier filter-to-filter fix. Track represented source
runes/gaps, project through char filters, publish exact provenance only for
unchanged candidates and a conservative span otherwise, and add these chains
with reset/reuse coverage.
##########
fe/fe-core/src/main/java/org/apache/doris/analysis/invertedindex/AnalyzerIdentityBuilder.java:
##########
@@ -226,26 +704,374 @@ private static String resolveTokenFilterIdentity(String
filterList) {
* IMPORTANT: Order is preserved because filter order is semantically
significant.
*/
private static String resolveCharFilterIdentity(String filterList) {
+ return resolveCharFilterIdentity(filterList, false);
+ }
+
+ private static String resolveCharFilterIdentity(String filterList, boolean
lowercaseDownstream) {
+ ArrayDeque<String> identities = new ArrayDeque<>();
+ walkCharFilters(filterList, lowercaseDownstream, identities);
+ return String.join(",", identities);
+ }
+
+ /**
+ * Resolve the chain from its last filter to its first, collecting
identities, and return the
+ * case-folding context that a filter placed in front of the chain would
run in.
+ */
+ private static boolean[] walkCharFilters(
+ String filterList, boolean lowercaseDownstream, Deque<String>
identities) {
+ boolean[] foldBlockedBytes = lowercaseDownstream ? new boolean[256] :
null;
if (Strings.isNullOrEmpty(filterList)) {
- return "";
+ return foldBlockedBytes;
}
- StringBuilder sb = new StringBuilder();
String[] filters = filterList.split(",\\s*");
// DO NOT sort - filter order is semantically significant
- for (int i = 0; i < filters.length; i++) {
- String filter = filters[i].trim();
- if (i > 0) {
- sb.append(",");
+ for (int i = filters.length - 1; i >= 0; --i) {
+ String filterName = filters[i].trim();
+ String filter = resolveComponentIdentity(
+ filterName, IndexPolicyTypeEnum.CHAR_FILTER,
foldBlockedBytes);
+ if (Strings.isNullOrEmpty(filter)) {
+ continue;
}
+ identities.addFirst(filter);
+ foldBlockedBytes = foldBlockedBytesBefore(filterName,
foldBlockedBytes);
+ }
+ return foldBlockedBytes;
+ }
- if (IndexPolicy.BUILTIN_CHAR_FILTERS.contains(filter)) {
- sb.append(filter);
- } else {
- sb.append(resolveComponentIdentity(filter,
IndexPolicyTypeEnum.CHAR_FILTER));
+ /**
+ * Context for the filter that runs before this one: a case fold starts a
fresh context, a
+ * char_replace filter adds the bytes it rewrites, and any other filter
ends the context.
+ */
+ private static boolean[] foldBlockedBytesBefore(String filterName,
boolean[] foldBlockedBytes) {
+ if (isCaseFoldingCharFilter(filterName)) {
+ return new boolean[256];
+ }
+ if (foldBlockedBytes == null) {
+ return null;
+ }
+ boolean[] sourceBytes = charReplaceSourceBytes(filterName);
+ if (sourceBytes == null) {
+ return null;
+ }
+ for (int i = 0; i < foldBlockedBytes.length; ++i) {
+ foldBlockedBytes[i] |= sourceBytes[i];
+ }
+ return foldBlockedBytes;
+ }
+
+ /** Bytes a named char_replace filter rewrites, or null for any other
filter. */
+ private static boolean[] charReplaceSourceBytes(String filterName) {
+ IndexPolicy policy = findPolicy(filterName,
IndexPolicyTypeEnum.CHAR_FILTER);
+ if (policy == null || policy.isInvalid() || policy.getProperties() ==
null) {
+ return null;
+ }
+ Map<String, String> properties = policy.getProperties();
+ String type = normalizeBuiltinComponentName(
+ properties.get(IndexPolicy.PROP_TYPE),
IndexPolicyTypeEnum.CHAR_FILTER);
+ String pattern = properties.get("pattern");
+ if (!"char_replace".equals(type) || pattern == null) {
+ return null;
+ }
+ boolean[] sourceBytes = new boolean[256];
+ for (int i = 0; i < pattern.length(); ++i) {
Review Comment:
[P1] Build the fold blocker from the effective replacement map, not the raw
pattern. This loop marks the replacement byte itself as rewritten even though
`canonicalizeCharReplacePattern` correctly drops that no-op entry. For
`lower_a={A->a}`, `x_to_A={pattern=Ax,replacement=A}`, and
`fold={icu_normalizer}`, `lower_a,x_to_A,fold` and `x_to_A,fold` are equivalent
for every byte, but raw `A` blocks elimination of `lower_a` and lets the
aliases pass CREATE/ALTER. This is a residual case beyond the earlier
non-interacting-filter fix. Exclude pattern bytes equal to the replacement and
add the mixed no-op pattern case to identity and DDL coverage.
##########
be/src/storage/index/inverted/token_filter/word_delimiter_filter.h:
##########
@@ -35,6 +36,14 @@ class WordDelimiterFilter : public DorisTokenFilter {
Token* next(Token* t) override;
void reset() override;
+ std::span<const int32_t> get_source_byte_offsets() const override {
Review Comment:
[P1] Publish a conservative span when WordDelimiter cannot slice an exact
map. If upstream provenance is empty, these getters stay empty and the
inherited conservative getter only delegates upstream, even though
WordDelimiter may have removed bytes. With raw `0xff 61`, Keyword ->
WordDelimiter emits `a` from source `[1,2)` but keeps `[0,2)` and no
provenance; a following offset-aware Pinyin filter falls back to term length
and reports `[0,1)`, the malformed byte. Snapshot an upstream/token span and
publish it for every generated part/concatenation lacking an exact slice, with
malformed leading/interior and reset/reuse coverage.
##########
fe/fe-core/src/main/java/org/apache/doris/analysis/invertedindex/AnalyzerIdentityBuilder.java:
##########
@@ -150,49 +221,458 @@ private static String
buildIdentityFromPolicyProperties(IndexPolicyTypeEnum type
* Resolve a component (tokenizer) to its identity.
*/
private static String resolveComponentIdentity(String name,
IndexPolicyTypeEnum expectedType) {
+ return resolveComponentIdentity(name, expectedType, null);
+ }
+
+ /**
+ * {@code foldBlockedBytes} is the case-folding context of a char filter:
null without a
+ * downstream fold, otherwise the bytes that filters between this one and
the fold rewrite.
+ */
+ private static String resolveComponentIdentity(
+ String name, IndexPolicyTypeEnum expectedType, boolean[]
foldBlockedBytes) {
if (Strings.isNullOrEmpty(name)) {
return "";
}
- // Check if it's a built-in component
- if (expectedType == IndexPolicyTypeEnum.TOKENIZER
- && IndexPolicy.BUILTIN_TOKENIZERS.contains(name)) {
- return name;
+ // Existing named policies take precedence over built-ins for upgrade
compatibility.
+ try {
+ Env env = Env.getCurrentEnv();
+ if (env != null && env.getIndexPolicyMgr() != null) {
+ IndexPolicy policy =
env.getIndexPolicyMgr().getPolicyByName(name);
+ if (policy != null && policy.getType() == expectedType) {
+ if (policy.isInvalid()) {
+ return "invalid-policy:" + policy.getId() + ":" +
policy.getName();
+ }
+ Map<String, String> props = policy.getProperties();
+ if (props != null && !props.isEmpty()) {
+ TreeMap<String, String> sortedProps = new
TreeMap<>(props);
+ String type = sortedProps.get(IndexPolicy.PROP_TYPE);
+ String normalizedType =
normalizeBuiltinComponentName(type, expectedType);
+ if (normalizedType != null) {
+ if ("empty".equals(normalizedType)) {
+ return "";
+ }
+ sortedProps.put(IndexPolicy.PROP_TYPE,
normalizedType);
+ canonicalizeEffectiveComponentProperties(
+ sortedProps, normalizedType, expectedType);
+ if (sortedProps.size() == 1) {
+ return normalizedType;
+ }
+ }
+ if (expectedType == IndexPolicyTypeEnum.TOKENIZER
+ &&
"ngram".equals(sortedProps.get(IndexPolicy.PROP_TYPE))) {
+ // This setting only limits policy creation; it
does not change emitted tokens.
+ sortedProps.remove(PROP_MAX_NGRAM_DIFF);
+ }
+ if (expectedType == IndexPolicyTypeEnum.CHAR_FILTER
+ &&
"char_replace".equals(sortedProps.get(IndexPolicy.PROP_TYPE))) {
+ String replacement =
sortedProps.getOrDefault("replacement", " ");
+ String pattern = canonicalizeCharReplacePattern(
+ sortedProps.get("pattern"), replacement,
foldBlockedBytes);
+ if (pattern.isEmpty()) {
+ return "";
+ }
+ sortedProps.put("pattern", pattern);
+ sortedProps.put("replacement", replacement);
+ }
+ if (normalizedType != null && sortedProps.size() == 1)
{
+ return normalizedType;
+ }
+ return sortedProps.toString();
+ }
+ }
+ }
+ } catch (RuntimeException e) {
+ // Fall through to built-in resolution or the original name.
+ }
+
+ String normalizedName = normalizeBuiltinComponentName(name,
expectedType);
+ return "empty".equals(normalizedName) ? "" : normalizedName == null ?
name : normalizedName;
+ }
+
+ private static void canonicalizeEffectiveComponentProperties(
+ TreeMap<String, String> properties, String type,
IndexPolicyTypeEnum expectedType) {
+ if ("pinyin".equals(type)) {
+ removeBooleanDefaults(properties, true,
+ "keep_first_letter", "keep_full_pinyin",
"keep_none_chinese",
+ "keep_none_chinese_together",
"keep_none_chinese_in_first_letter",
+ "lowercase", "trim_whitespace", "ignore_pinyin_offset",
+ "none_chinese_pinyin_tokenize");
+ removeBooleanDefaults(properties, false,
+ "keep_separate_first_letter", "keep_joined_full_pinyin",
"keep_original",
+ "keep_none_chinese_in_joined_full_pinyin",
"remove_duplicated_term",
+ "fixed_pinyin_offset", "keep_separate_chinese");
+ removeIntegerDefault(properties, "limit_first_letter_length", 16);
+ canonicalizePinyinDependencies(properties, expectedType);
+ return;
+ }
+
+ if (expectedType == IndexPolicyTypeEnum.TOKEN_FILTER) {
+ if ("asciifolding".equals(type)) {
+ removeBooleanDefaults(properties, false, "preserve_original");
+ } else if ("word_delimiter".equals(type)) {
+ removeBooleanDefaults(properties, true, "generate_word_parts",
"generate_number_parts",
+ "split_on_case_change", "split_on_numerics",
"stem_english_possessive");
+ removeBooleanDefaults(properties, false, "catenate_words",
"catenate_numbers",
+ "catenate_all", "preserve_original");
+ canonicalizeWordSet(properties, "protected_words");
+ canonicalizeTypeTable(properties);
+ } else if ("icu_normalizer".equals(type)) {
+ canonicalizeIcuNormalizerDefaults(properties, false);
+ }
+ return;
}
- // For custom component, get its properties
+ if (expectedType == IndexPolicyTypeEnum.CHAR_FILTER) {
+ if ("icu_normalizer".equals(type)) {
+ canonicalizeIcuNormalizerDefaults(properties, true);
+ }
+ return;
+ }
+
+ if (expectedType != IndexPolicyTypeEnum.TOKENIZER) {
+ return;
+ }
+ switch (type) {
+ case "ngram":
+ case "edge_ngram":
+ removeIntegerDefault(properties, "min_gram", 1);
+ removeIntegerDefault(properties, "max_gram", 2);
+ canonicalizeWordSet(properties, "token_chars");
+ canonicalizeCustomTokenChars(properties);
+ break;
+ case "standard":
+ removeIntegerDefault(properties, "max_token_length", 255);
+ break;
+ case "char_group":
+ removeIntegerDefault(properties, "max_token_length", 255);
+ canonicalizeTokenizeOnChars(properties);
Review Comment:
[P1] Remove CharGroup literals already absorbed by an enabled category. BE
checks category predicates before the literal set, so
`tokenize_on_chars=[letter]` and `[letter],[A]` split every input identically
and produce the same terms, offsets, and provenance. Sorting/deduplicating the
raw entries still leaves `A` in one identity, allowing differently named
aliases through CREATE/ALTER. Canonicalize literals against the effective
category predicates and add this alias case plus an uncovered-literal negative.
##########
fe/fe-core/src/main/java/org/apache/doris/analysis/invertedindex/AnalyzerIdentityBuilder.java:
##########
@@ -150,49 +221,458 @@ private static String
buildIdentityFromPolicyProperties(IndexPolicyTypeEnum type
* Resolve a component (tokenizer) to its identity.
*/
private static String resolveComponentIdentity(String name,
IndexPolicyTypeEnum expectedType) {
+ return resolveComponentIdentity(name, expectedType, null);
+ }
+
+ /**
+ * {@code foldBlockedBytes} is the case-folding context of a char filter:
null without a
+ * downstream fold, otherwise the bytes that filters between this one and
the fold rewrite.
+ */
+ private static String resolveComponentIdentity(
+ String name, IndexPolicyTypeEnum expectedType, boolean[]
foldBlockedBytes) {
if (Strings.isNullOrEmpty(name)) {
return "";
}
- // Check if it's a built-in component
- if (expectedType == IndexPolicyTypeEnum.TOKENIZER
- && IndexPolicy.BUILTIN_TOKENIZERS.contains(name)) {
- return name;
+ // Existing named policies take precedence over built-ins for upgrade
compatibility.
+ try {
+ Env env = Env.getCurrentEnv();
+ if (env != null && env.getIndexPolicyMgr() != null) {
+ IndexPolicy policy =
env.getIndexPolicyMgr().getPolicyByName(name);
+ if (policy != null && policy.getType() == expectedType) {
+ if (policy.isInvalid()) {
+ return "invalid-policy:" + policy.getId() + ":" +
policy.getName();
+ }
+ Map<String, String> props = policy.getProperties();
+ if (props != null && !props.isEmpty()) {
+ TreeMap<String, String> sortedProps = new
TreeMap<>(props);
+ String type = sortedProps.get(IndexPolicy.PROP_TYPE);
+ String normalizedType =
normalizeBuiltinComponentName(type, expectedType);
+ if (normalizedType != null) {
+ if ("empty".equals(normalizedType)) {
+ return "";
+ }
+ sortedProps.put(IndexPolicy.PROP_TYPE,
normalizedType);
+ canonicalizeEffectiveComponentProperties(
+ sortedProps, normalizedType, expectedType);
+ if (sortedProps.size() == 1) {
+ return normalizedType;
+ }
+ }
+ if (expectedType == IndexPolicyTypeEnum.TOKENIZER
+ &&
"ngram".equals(sortedProps.get(IndexPolicy.PROP_TYPE))) {
+ // This setting only limits policy creation; it
does not change emitted tokens.
+ sortedProps.remove(PROP_MAX_NGRAM_DIFF);
+ }
+ if (expectedType == IndexPolicyTypeEnum.CHAR_FILTER
+ &&
"char_replace".equals(sortedProps.get(IndexPolicy.PROP_TYPE))) {
+ String replacement =
sortedProps.getOrDefault("replacement", " ");
+ String pattern = canonicalizeCharReplacePattern(
+ sortedProps.get("pattern"), replacement,
foldBlockedBytes);
+ if (pattern.isEmpty()) {
+ return "";
+ }
+ sortedProps.put("pattern", pattern);
+ sortedProps.put("replacement", replacement);
+ }
+ if (normalizedType != null && sortedProps.size() == 1)
{
+ return normalizedType;
+ }
+ return sortedProps.toString();
+ }
+ }
+ }
+ } catch (RuntimeException e) {
+ // Fall through to built-in resolution or the original name.
+ }
+
+ String normalizedName = normalizeBuiltinComponentName(name,
expectedType);
+ return "empty".equals(normalizedName) ? "" : normalizedName == null ?
name : normalizedName;
+ }
+
+ private static void canonicalizeEffectiveComponentProperties(
+ TreeMap<String, String> properties, String type,
IndexPolicyTypeEnum expectedType) {
+ if ("pinyin".equals(type)) {
+ removeBooleanDefaults(properties, true,
+ "keep_first_letter", "keep_full_pinyin",
"keep_none_chinese",
+ "keep_none_chinese_together",
"keep_none_chinese_in_first_letter",
+ "lowercase", "trim_whitespace", "ignore_pinyin_offset",
+ "none_chinese_pinyin_tokenize");
+ removeBooleanDefaults(properties, false,
+ "keep_separate_first_letter", "keep_joined_full_pinyin",
"keep_original",
+ "keep_none_chinese_in_joined_full_pinyin",
"remove_duplicated_term",
Review Comment:
[P1] Drop `remove_duplicated_term` when the effective Pinyin output shape
has cardinality at most one. With first/full/separate/original/non-Chinese
output disabled and only joined-full-pinyin enabled, Chinese input emits one
joined candidate; other input emits none in the tokenizer or one fallback
original in the token filter. The dedup flag cannot affect terms, positions,
offsets, or provenance, yet retaining it gives differently named aliases
different identities and lets both through CREATE/ALTER. Derive whether
multiple candidates are possible, remove the flag when not, and keep a
multi-output negative case.
##########
be/src/storage/index/inverted/token_filter/ascii_folding_filter.cpp:
##########
@@ -42,7 +64,15 @@ Token* ASCIIFoldingFilter::next(Token* t) {
continue;
}
if (c >= 0x0080) {
+ const int32_t input_runes =
count_utf8_runes(std::string_view(buffer, length));
fold_to_ascii(buffer, length);
+ _rune_count_changed =
Review Comment:
[P1] Treat malformed input that folding rewrites as provenance-changing.
`count_utf8_runes()` returns `-1`, but this condition then leaves
`_rune_count_changed=false` even though `fold_to_ascii()` skips malformed bytes
and continues. For raw `0xff C3 86`, Keyword -> asciifolding emits `AE` with no
map/span; a following offset-aware Pinyin filter reports `[0,2)`, assigning `A`
to the discarded byte and ending `E` inside the original `Æ` at `[1,3)`. Reject
malformed input or publish the whole upstream token span whenever the input
count is invalid, and cover preserve-original plus reset/reuse.
##########
fe/fe-core/src/main/java/org/apache/doris/analysis/invertedindex/AnalyzerIdentityBuilder.java:
##########
@@ -150,49 +221,458 @@ private static String
buildIdentityFromPolicyProperties(IndexPolicyTypeEnum type
* Resolve a component (tokenizer) to its identity.
*/
private static String resolveComponentIdentity(String name,
IndexPolicyTypeEnum expectedType) {
+ return resolveComponentIdentity(name, expectedType, null);
+ }
+
+ /**
+ * {@code foldBlockedBytes} is the case-folding context of a char filter:
null without a
+ * downstream fold, otherwise the bytes that filters between this one and
the fold rewrite.
+ */
+ private static String resolveComponentIdentity(
+ String name, IndexPolicyTypeEnum expectedType, boolean[]
foldBlockedBytes) {
if (Strings.isNullOrEmpty(name)) {
return "";
}
- // Check if it's a built-in component
- if (expectedType == IndexPolicyTypeEnum.TOKENIZER
- && IndexPolicy.BUILTIN_TOKENIZERS.contains(name)) {
- return name;
+ // Existing named policies take precedence over built-ins for upgrade
compatibility.
+ try {
+ Env env = Env.getCurrentEnv();
+ if (env != null && env.getIndexPolicyMgr() != null) {
+ IndexPolicy policy =
env.getIndexPolicyMgr().getPolicyByName(name);
+ if (policy != null && policy.getType() == expectedType) {
+ if (policy.isInvalid()) {
+ return "invalid-policy:" + policy.getId() + ":" +
policy.getName();
+ }
+ Map<String, String> props = policy.getProperties();
+ if (props != null && !props.isEmpty()) {
+ TreeMap<String, String> sortedProps = new
TreeMap<>(props);
+ String type = sortedProps.get(IndexPolicy.PROP_TYPE);
+ String normalizedType =
normalizeBuiltinComponentName(type, expectedType);
+ if (normalizedType != null) {
+ if ("empty".equals(normalizedType)) {
+ return "";
+ }
+ sortedProps.put(IndexPolicy.PROP_TYPE,
normalizedType);
+ canonicalizeEffectiveComponentProperties(
+ sortedProps, normalizedType, expectedType);
+ if (sortedProps.size() == 1) {
+ return normalizedType;
+ }
+ }
+ if (expectedType == IndexPolicyTypeEnum.TOKENIZER
+ &&
"ngram".equals(sortedProps.get(IndexPolicy.PROP_TYPE))) {
+ // This setting only limits policy creation; it
does not change emitted tokens.
+ sortedProps.remove(PROP_MAX_NGRAM_DIFF);
+ }
+ if (expectedType == IndexPolicyTypeEnum.CHAR_FILTER
+ &&
"char_replace".equals(sortedProps.get(IndexPolicy.PROP_TYPE))) {
+ String replacement =
sortedProps.getOrDefault("replacement", " ");
+ String pattern = canonicalizeCharReplacePattern(
+ sortedProps.get("pattern"), replacement,
foldBlockedBytes);
+ if (pattern.isEmpty()) {
+ return "";
+ }
+ sortedProps.put("pattern", pattern);
+ sortedProps.put("replacement", replacement);
+ }
+ if (normalizedType != null && sortedProps.size() == 1)
{
+ return normalizedType;
+ }
+ return sortedProps.toString();
+ }
+ }
+ }
+ } catch (RuntimeException e) {
+ // Fall through to built-in resolution or the original name.
+ }
+
+ String normalizedName = normalizeBuiltinComponentName(name,
expectedType);
+ return "empty".equals(normalizedName) ? "" : normalizedName == null ?
name : normalizedName;
+ }
+
+ private static void canonicalizeEffectiveComponentProperties(
+ TreeMap<String, String> properties, String type,
IndexPolicyTypeEnum expectedType) {
+ if ("pinyin".equals(type)) {
+ removeBooleanDefaults(properties, true,
+ "keep_first_letter", "keep_full_pinyin",
"keep_none_chinese",
+ "keep_none_chinese_together",
"keep_none_chinese_in_first_letter",
+ "lowercase", "trim_whitespace", "ignore_pinyin_offset",
+ "none_chinese_pinyin_tokenize");
+ removeBooleanDefaults(properties, false,
+ "keep_separate_first_letter", "keep_joined_full_pinyin",
"keep_original",
+ "keep_none_chinese_in_joined_full_pinyin",
"remove_duplicated_term",
+ "fixed_pinyin_offset", "keep_separate_chinese");
+ removeIntegerDefault(properties, "limit_first_letter_length", 16);
+ canonicalizePinyinDependencies(properties, expectedType);
+ return;
+ }
+
+ if (expectedType == IndexPolicyTypeEnum.TOKEN_FILTER) {
+ if ("asciifolding".equals(type)) {
+ removeBooleanDefaults(properties, false, "preserve_original");
+ } else if ("word_delimiter".equals(type)) {
+ removeBooleanDefaults(properties, true, "generate_word_parts",
"generate_number_parts",
+ "split_on_case_change", "split_on_numerics",
"stem_english_possessive");
+ removeBooleanDefaults(properties, false, "catenate_words",
"catenate_numbers",
+ "catenate_all", "preserve_original");
+ canonicalizeWordSet(properties, "protected_words");
+ canonicalizeTypeTable(properties);
+ } else if ("icu_normalizer".equals(type)) {
+ canonicalizeIcuNormalizerDefaults(properties, false);
+ }
+ return;
}
- // For custom component, get its properties
+ if (expectedType == IndexPolicyTypeEnum.CHAR_FILTER) {
+ if ("icu_normalizer".equals(type)) {
+ canonicalizeIcuNormalizerDefaults(properties, true);
+ }
+ return;
+ }
+
+ if (expectedType != IndexPolicyTypeEnum.TOKENIZER) {
+ return;
+ }
+ switch (type) {
+ case "ngram":
+ case "edge_ngram":
+ removeIntegerDefault(properties, "min_gram", 1);
+ removeIntegerDefault(properties, "max_gram", 2);
+ canonicalizeWordSet(properties, "token_chars");
+ canonicalizeCustomTokenChars(properties);
+ break;
+ case "standard":
+ removeIntegerDefault(properties, "max_token_length", 255);
+ break;
+ case "char_group":
+ removeIntegerDefault(properties, "max_token_length", 255);
+ canonicalizeTokenizeOnChars(properties);
+ break;
+ case "keyword":
+ // BE only range-checks buffer_size; the emitted term is
always capped by a constant.
+ properties.remove("buffer_size");
+ break;
+ case "basic":
+ canonicalizeBasicExtraChars(properties);
+ break;
+ default:
+ break;
+ }
+ }
+
+ private static void removeBooleanDefaults(
+ TreeMap<String, String> properties, boolean defaultValue,
String... keys) {
+ for (String key : keys) {
+ String value = properties.get(key);
+ if (value == null || !("true".equalsIgnoreCase(value) ||
"false".equalsIgnoreCase(value))) {
+ continue;
+ }
+ boolean parsed = Boolean.parseBoolean(value);
+ if (parsed == defaultValue) {
+ properties.remove(key);
+ } else {
+ properties.put(key, Boolean.toString(parsed));
+ }
+ }
+ }
+
+ private static void removeIntegerDefault(
+ TreeMap<String, String> properties, String key, int defaultValue) {
+ String value = properties.get(key);
+ if (value == null) {
+ return;
+ }
try {
- Env env = Env.getCurrentEnv();
- if (env == null || env.getIndexPolicyMgr() == null) {
- return name;
+ int parsed = Integer.parseInt(value);
+ if (parsed == defaultValue) {
+ properties.remove(key);
+ } else {
+ properties.put(key, Integer.toString(parsed));
}
+ } catch (NumberFormatException e) {
+ // Invalid policies keep their original identity.
+ }
+ }
- IndexPolicy policy = env.getIndexPolicyMgr().getPolicyByName(name);
- if (policy == null || policy.getType() != expectedType) {
- return name;
+ private static void canonicalizeIcuNormalizerDefaults(
+ TreeMap<String, String> properties, boolean hasMode) {
+ String name = properties.get("name");
+ if (name != null) {
+ String normalizedName = name.trim().toLowerCase(Locale.ROOT);
+ if ("nfkc_cf".equals(normalizedName)) {
+ properties.remove("name");
+ } else {
+ properties.put("name", normalizedName);
}
- if (policy.isInvalid()) {
- return "invalid-policy:" + policy.getId() + ":" +
policy.getName();
+ }
+ String filter = properties.get("unicode_set_filter");
+ if (filter != null && filter.isEmpty()) {
+ // BE treats an explicit empty string like an absent filter.
+ properties.remove("unicode_set_filter");
+ } else if (filter != null) {
+ try {
+ UnicodeSet unicodeSet = new UnicodeSet(filter);
+ if (unicodeSet.isEmpty()) {
+ properties.remove("unicode_set_filter");
+ } else {
+ properties.put("unicode_set_filter",
unicodeSet.toPattern(false));
+ }
+ } catch (IllegalArgumentException e) {
+ // Invalid policies keep their original identity.
}
+ }
+ if (hasMode) {
+ canonicalizeIcuNormalizerMode(properties);
+ }
+ }
- Map<String, String> props = policy.getProperties();
- if (props == null || props.isEmpty()) {
- return name;
+ private static void canonicalizeIcuNormalizerMode(TreeMap<String, String>
properties) {
+ removeStringDefault(properties, "mode", "compose");
+ if (!"decompose".equals(properties.get("mode"))) {
+ return;
+ }
+ // BE ignores mode for nfd/nfkd, and nfc/nfkc in decompose mode are
the same ICU instances.
+ String name = properties.get("name");
+ if ("nfc".equals(name) || "nfd".equals(name)) {
+ properties.put("name", "nfd");
+ properties.remove("mode");
+ } else if ("nfkc".equals(name) || "nfkd".equals(name)) {
+ properties.put("name", "nfkd");
+ properties.remove("mode");
+ }
+ }
+
+ // BE reads these settings as unordered sets of trimmed, non-empty words.
+ private static void canonicalizeWordSet(TreeMap<String, String>
properties, String key) {
+ String value = properties.get(key);
+ if (value == null) {
+ return;
+ }
+ TreeSet<String> words = new TreeSet<>();
+ for (String word : value.split(",")) {
+ String trimmed = trimAsciiWhitespace(word);
+ if (!trimmed.isEmpty()) {
+ words.add(trimmed);
}
+ }
+ if (words.isEmpty()) {
+ properties.remove(key);
+ } else {
+ properties.put(key, String.join(",", words));
+ }
+ }
+
+ // BE matches custom token characters as a code point set.
+ private static void canonicalizeCustomTokenChars(TreeMap<String, String>
properties) {
+ String value = properties.get("custom_token_chars");
+ if (value == null) {
+ return;
+ }
+ StringBuilder canonical = new StringBuilder();
+
value.codePoints().distinct().sorted().forEach(canonical::appendCodePoint);
+ properties.put("custom_token_chars", canonical.toString());
+ }
- // Build identity from sorted properties
- TreeMap<String, String> sortedProps = new TreeMap<>(props);
- if (expectedType == IndexPolicyTypeEnum.TOKENIZER
- && "ngram".equals(sortedProps.get(IndexPolicy.PROP_TYPE)))
{
- // This setting only limits policy creation; it does not
change emitted tokens.
- sortedProps.remove(PROP_MAX_NGRAM_DIFF);
+ // BE collects tokenize_on_chars entries into sets, so order and repeats
do not matter.
+ private static void canonicalizeTokenizeOnChars(TreeMap<String, String>
properties) {
+ List<String> entries =
parseEntryList(properties.get("tokenize_on_chars"));
+ if (entries == null) {
+ return;
+ }
+ putEntryList(properties, "tokenize_on_chars", new TreeSet<>(entries));
+ }
+
+ // BE builds a per-character type map where a later rule for the same
character wins.
+ private static void canonicalizeTypeTable(TreeMap<String, String>
properties) {
+ List<String> rules = parseEntryList(properties.get("type_table"));
+ if (rules == null) {
+ return;
+ }
+ TreeMap<Integer, String> types = new TreeMap<>();
+ for (String rule : rules) {
+ int arrow = rule.lastIndexOf("=>");
+ if (arrow < 0 || rule.indexOf('\n') >= 0 || rule.indexOf('\r') >=
0) {
+ return;
}
- return sortedProps.toString();
- } catch (RuntimeException e) {
- return name;
+ String character = trimAsciiWhitespace(rule.substring(0, arrow));
+ String type = trimAsciiWhitespace(rule.substring(arrow + 2));
+ // Escaped characters keep the original identity rather than
reproducing BE unescaping.
+ if (character.indexOf('\\') >= 0 || character.codePointCount(0,
character.length()) != 1
+ || !WORD_DELIMITER_TYPES.contains(type)) {
+ return;
+ }
+ types.put(character.codePointAt(0), type);
+ }
+ List<String> canonicalRules = new ArrayList<>();
+ for (Map.Entry<Integer, String> entry : types.entrySet()) {
+ canonicalRules.add(new String(Character.toChars(entry.getKey())) +
"=>" + entry.getValue());
Review Comment:
[P1] Remove `type_table` rules that merely restate WordDelimiter's inherited
Unicode type. BE initializes the custom table with
`WordDelimiterIterator::get_type()` and then applies overrides, so `a` is
already `LOWER`; analyzers with no table and with `[a => LOWER]` emit identical
terms, positions, and offsets. This method still serializes that redundant
rule, giving differently named aliases different identities and letting both
through CREATE/ALTER duplicate checks. This is residual to the order/last-rule
thread because the final map itself still contains a no-op override. Compare
final rules with BE's default classification, drop no-ops, and cover `[a =>
LOWER]` versus absent plus `[a => DIGIT]` as a negative.
##########
fe/fe-core/src/main/java/org/apache/doris/analysis/invertedindex/AnalyzerIdentityBuilder.java:
##########
@@ -226,26 +704,374 @@ private static String resolveTokenFilterIdentity(String
filterList) {
* IMPORTANT: Order is preserved because filter order is semantically
significant.
*/
private static String resolveCharFilterIdentity(String filterList) {
+ return resolveCharFilterIdentity(filterList, false);
+ }
+
+ private static String resolveCharFilterIdentity(String filterList, boolean
lowercaseDownstream) {
+ ArrayDeque<String> identities = new ArrayDeque<>();
+ walkCharFilters(filterList, lowercaseDownstream, identities);
+ return String.join(",", identities);
+ }
+
+ /**
+ * Resolve the chain from its last filter to its first, collecting
identities, and return the
+ * case-folding context that a filter placed in front of the chain would
run in.
+ */
+ private static boolean[] walkCharFilters(
+ String filterList, boolean lowercaseDownstream, Deque<String>
identities) {
+ boolean[] foldBlockedBytes = lowercaseDownstream ? new boolean[256] :
null;
if (Strings.isNullOrEmpty(filterList)) {
- return "";
+ return foldBlockedBytes;
}
- StringBuilder sb = new StringBuilder();
String[] filters = filterList.split(",\\s*");
// DO NOT sort - filter order is semantically significant
- for (int i = 0; i < filters.length; i++) {
- String filter = filters[i].trim();
- if (i > 0) {
- sb.append(",");
+ for (int i = filters.length - 1; i >= 0; --i) {
+ String filterName = filters[i].trim();
+ String filter = resolveComponentIdentity(
+ filterName, IndexPolicyTypeEnum.CHAR_FILTER,
foldBlockedBytes);
+ if (Strings.isNullOrEmpty(filter)) {
+ continue;
}
+ identities.addFirst(filter);
+ foldBlockedBytes = foldBlockedBytesBefore(filterName,
foldBlockedBytes);
+ }
+ return foldBlockedBytes;
+ }
- if (IndexPolicy.BUILTIN_CHAR_FILTERS.contains(filter)) {
- sb.append(filter);
- } else {
- sb.append(resolveComponentIdentity(filter,
IndexPolicyTypeEnum.CHAR_FILTER));
+ /**
+ * Context for the filter that runs before this one: a case fold starts a
fresh context, a
+ * char_replace filter adds the bytes it rewrites, and any other filter
ends the context.
+ */
+ private static boolean[] foldBlockedBytesBefore(String filterName,
boolean[] foldBlockedBytes) {
+ if (isCaseFoldingCharFilter(filterName)) {
+ return new boolean[256];
+ }
+ if (foldBlockedBytes == null) {
+ return null;
+ }
+ boolean[] sourceBytes = charReplaceSourceBytes(filterName);
+ if (sourceBytes == null) {
+ return null;
+ }
+ for (int i = 0; i < foldBlockedBytes.length; ++i) {
+ foldBlockedBytes[i] |= sourceBytes[i];
+ }
+ return foldBlockedBytes;
+ }
+
+ /** Bytes a named char_replace filter rewrites, or null for any other
filter. */
+ private static boolean[] charReplaceSourceBytes(String filterName) {
+ IndexPolicy policy = findPolicy(filterName,
IndexPolicyTypeEnum.CHAR_FILTER);
+ if (policy == null || policy.isInvalid() || policy.getProperties() ==
null) {
+ return null;
+ }
+ Map<String, String> properties = policy.getProperties();
+ String type = normalizeBuiltinComponentName(
+ properties.get(IndexPolicy.PROP_TYPE),
IndexPolicyTypeEnum.CHAR_FILTER);
+ String pattern = properties.get("pattern");
+ if (!"char_replace".equals(type) || pattern == null) {
+ return null;
+ }
+ boolean[] sourceBytes = new boolean[256];
+ for (int i = 0; i < pattern.length(); ++i) {
+ char patternByte = pattern.charAt(i);
+ if (patternByte < sourceBytes.length) {
+ sourceBytes[patternByte] = true;
}
}
- return sb.toString();
+ return sourceBytes;
+ }
+
+ /** The named policy when one exists with the expected type, or null. */
+ private static IndexPolicy findPolicy(String name, IndexPolicyTypeEnum
expectedType) {
+ if (Strings.isNullOrEmpty(name)) {
+ return null;
+ }
+ try {
+ Env env = Env.getCurrentEnv();
+ if (env != null && env.getIndexPolicyMgr() != null) {
+ IndexPolicy policy =
env.getIndexPolicyMgr().getPolicyByName(name);
+ if (policy != null && policy.getType() == expectedType) {
+ return policy;
+ }
+ }
+ } catch (RuntimeException e) {
+ // Treat lookup failures as an unknown policy.
+ }
+ return null;
+ }
+
+ private static boolean isCaseFoldingCharFilter(String name) {
+ if (Strings.isNullOrEmpty(name)) {
+ return false;
+ }
+
+ try {
+ Env env = Env.getCurrentEnv();
+ if (env != null && env.getIndexPolicyMgr() != null) {
+ IndexPolicy policy =
env.getIndexPolicyMgr().getPolicyByName(name);
+ if (policy != null && policy.getType() ==
IndexPolicyTypeEnum.CHAR_FILTER) {
+ if (policy.isInvalid()) {
+ return false;
+ }
+ Map<String, String> properties = policy.getProperties();
+ if (properties != null && !properties.isEmpty()) {
+ String type = normalizeBuiltinComponentName(
+ properties.get(IndexPolicy.PROP_TYPE),
IndexPolicyTypeEnum.CHAR_FILTER);
+ return "icu_normalizer".equals(type) &&
isCaseFoldingIcuNormalizer(properties);
+ }
+ }
+ }
+ } catch (RuntimeException e) {
+ // Fall through to built-in resolution.
+ }
+
+ return "icu_normalizer".equals(
+ normalizeBuiltinComponentName(name,
IndexPolicyTypeEnum.CHAR_FILTER));
+ }
+
+ /** Whether an icu_normalizer component folds case: the default nfkc_cf
form over every code point. */
+ private static boolean isCaseFoldingIcuNormalizer(Map<String, String>
properties) {
Review Comment:
[P1] Preserve enough fold context for non-empty ICU sets to prove the
specific mapping being canonicalized. BE's `FilteredNormalizer2(nfkc_cf, [A])`
maps `A` to `a`, while input `a` is already unchanged outside the set, so
`keyword -> icu_normalizer(unicode_set_filter=[A])` emits the same stream with
or without outer `char_replace A->a`. Requiring an unfiltered normalizer here
retains the outer suffix and lets equivalent aliases pass CREATE/ALTER. This is
distinct from the parsed-empty-set thread: the effective set is non-empty and
contains the exact uppercase source. Carry a pattern-aware set/fold context,
with `[A]` positive and `[B]` negative identity plus DDL coverage.
##########
fe/fe-core/src/main/java/org/apache/doris/analysis/invertedindex/AnalyzerIdentityBuilder.java:
##########
@@ -150,49 +221,458 @@ private static String
buildIdentityFromPolicyProperties(IndexPolicyTypeEnum type
* Resolve a component (tokenizer) to its identity.
*/
private static String resolveComponentIdentity(String name,
IndexPolicyTypeEnum expectedType) {
+ return resolveComponentIdentity(name, expectedType, null);
+ }
+
+ /**
+ * {@code foldBlockedBytes} is the case-folding context of a char filter:
null without a
+ * downstream fold, otherwise the bytes that filters between this one and
the fold rewrite.
+ */
+ private static String resolveComponentIdentity(
+ String name, IndexPolicyTypeEnum expectedType, boolean[]
foldBlockedBytes) {
if (Strings.isNullOrEmpty(name)) {
return "";
}
- // Check if it's a built-in component
- if (expectedType == IndexPolicyTypeEnum.TOKENIZER
- && IndexPolicy.BUILTIN_TOKENIZERS.contains(name)) {
- return name;
+ // Existing named policies take precedence over built-ins for upgrade
compatibility.
+ try {
+ Env env = Env.getCurrentEnv();
+ if (env != null && env.getIndexPolicyMgr() != null) {
+ IndexPolicy policy =
env.getIndexPolicyMgr().getPolicyByName(name);
+ if (policy != null && policy.getType() == expectedType) {
+ if (policy.isInvalid()) {
+ return "invalid-policy:" + policy.getId() + ":" +
policy.getName();
+ }
+ Map<String, String> props = policy.getProperties();
+ if (props != null && !props.isEmpty()) {
+ TreeMap<String, String> sortedProps = new
TreeMap<>(props);
+ String type = sortedProps.get(IndexPolicy.PROP_TYPE);
+ String normalizedType =
normalizeBuiltinComponentName(type, expectedType);
+ if (normalizedType != null) {
+ if ("empty".equals(normalizedType)) {
+ return "";
+ }
+ sortedProps.put(IndexPolicy.PROP_TYPE,
normalizedType);
+ canonicalizeEffectiveComponentProperties(
+ sortedProps, normalizedType, expectedType);
+ if (sortedProps.size() == 1) {
+ return normalizedType;
+ }
+ }
+ if (expectedType == IndexPolicyTypeEnum.TOKENIZER
+ &&
"ngram".equals(sortedProps.get(IndexPolicy.PROP_TYPE))) {
+ // This setting only limits policy creation; it
does not change emitted tokens.
+ sortedProps.remove(PROP_MAX_NGRAM_DIFF);
+ }
+ if (expectedType == IndexPolicyTypeEnum.CHAR_FILTER
+ &&
"char_replace".equals(sortedProps.get(IndexPolicy.PROP_TYPE))) {
+ String replacement =
sortedProps.getOrDefault("replacement", " ");
+ String pattern = canonicalizeCharReplacePattern(
+ sortedProps.get("pattern"), replacement,
foldBlockedBytes);
+ if (pattern.isEmpty()) {
+ return "";
+ }
+ sortedProps.put("pattern", pattern);
+ sortedProps.put("replacement", replacement);
+ }
+ if (normalizedType != null && sortedProps.size() == 1)
{
+ return normalizedType;
+ }
+ return sortedProps.toString();
+ }
+ }
+ }
+ } catch (RuntimeException e) {
+ // Fall through to built-in resolution or the original name.
+ }
+
+ String normalizedName = normalizeBuiltinComponentName(name,
expectedType);
+ return "empty".equals(normalizedName) ? "" : normalizedName == null ?
name : normalizedName;
+ }
+
+ private static void canonicalizeEffectiveComponentProperties(
+ TreeMap<String, String> properties, String type,
IndexPolicyTypeEnum expectedType) {
+ if ("pinyin".equals(type)) {
+ removeBooleanDefaults(properties, true,
+ "keep_first_letter", "keep_full_pinyin",
"keep_none_chinese",
+ "keep_none_chinese_together",
"keep_none_chinese_in_first_letter",
+ "lowercase", "trim_whitespace", "ignore_pinyin_offset",
+ "none_chinese_pinyin_tokenize");
+ removeBooleanDefaults(properties, false,
+ "keep_separate_first_letter", "keep_joined_full_pinyin",
"keep_original",
+ "keep_none_chinese_in_joined_full_pinyin",
"remove_duplicated_term",
+ "fixed_pinyin_offset", "keep_separate_chinese");
+ removeIntegerDefault(properties, "limit_first_letter_length", 16);
+ canonicalizePinyinDependencies(properties, expectedType);
+ return;
+ }
+
+ if (expectedType == IndexPolicyTypeEnum.TOKEN_FILTER) {
+ if ("asciifolding".equals(type)) {
+ removeBooleanDefaults(properties, false, "preserve_original");
+ } else if ("word_delimiter".equals(type)) {
+ removeBooleanDefaults(properties, true, "generate_word_parts",
"generate_number_parts",
+ "split_on_case_change", "split_on_numerics",
"stem_english_possessive");
+ removeBooleanDefaults(properties, false, "catenate_words",
"catenate_numbers",
+ "catenate_all", "preserve_original");
+ canonicalizeWordSet(properties, "protected_words");
+ canonicalizeTypeTable(properties);
+ } else if ("icu_normalizer".equals(type)) {
+ canonicalizeIcuNormalizerDefaults(properties, false);
+ }
+ return;
}
- // For custom component, get its properties
+ if (expectedType == IndexPolicyTypeEnum.CHAR_FILTER) {
+ if ("icu_normalizer".equals(type)) {
+ canonicalizeIcuNormalizerDefaults(properties, true);
+ }
+ return;
+ }
+
+ if (expectedType != IndexPolicyTypeEnum.TOKENIZER) {
+ return;
+ }
+ switch (type) {
+ case "ngram":
+ case "edge_ngram":
+ removeIntegerDefault(properties, "min_gram", 1);
+ removeIntegerDefault(properties, "max_gram", 2);
+ canonicalizeWordSet(properties, "token_chars");
+ canonicalizeCustomTokenChars(properties);
+ break;
+ case "standard":
+ removeIntegerDefault(properties, "max_token_length", 255);
+ break;
+ case "char_group":
+ removeIntegerDefault(properties, "max_token_length", 255);
+ canonicalizeTokenizeOnChars(properties);
+ break;
+ case "keyword":
+ // BE only range-checks buffer_size; the emitted term is
always capped by a constant.
+ properties.remove("buffer_size");
+ break;
+ case "basic":
+ canonicalizeBasicExtraChars(properties);
+ break;
+ default:
+ break;
+ }
+ }
+
+ private static void removeBooleanDefaults(
+ TreeMap<String, String> properties, boolean defaultValue,
String... keys) {
+ for (String key : keys) {
+ String value = properties.get(key);
+ if (value == null || !("true".equalsIgnoreCase(value) ||
"false".equalsIgnoreCase(value))) {
+ continue;
+ }
+ boolean parsed = Boolean.parseBoolean(value);
+ if (parsed == defaultValue) {
+ properties.remove(key);
+ } else {
+ properties.put(key, Boolean.toString(parsed));
+ }
+ }
+ }
+
+ private static void removeIntegerDefault(
+ TreeMap<String, String> properties, String key, int defaultValue) {
+ String value = properties.get(key);
+ if (value == null) {
+ return;
+ }
try {
- Env env = Env.getCurrentEnv();
- if (env == null || env.getIndexPolicyMgr() == null) {
- return name;
+ int parsed = Integer.parseInt(value);
+ if (parsed == defaultValue) {
+ properties.remove(key);
+ } else {
+ properties.put(key, Integer.toString(parsed));
}
+ } catch (NumberFormatException e) {
+ // Invalid policies keep their original identity.
+ }
+ }
- IndexPolicy policy = env.getIndexPolicyMgr().getPolicyByName(name);
- if (policy == null || policy.getType() != expectedType) {
- return name;
+ private static void canonicalizeIcuNormalizerDefaults(
+ TreeMap<String, String> properties, boolean hasMode) {
+ String name = properties.get("name");
+ if (name != null) {
+ String normalizedName = name.trim().toLowerCase(Locale.ROOT);
+ if ("nfkc_cf".equals(normalizedName)) {
+ properties.remove("name");
+ } else {
+ properties.put("name", normalizedName);
}
- if (policy.isInvalid()) {
- return "invalid-policy:" + policy.getId() + ":" +
policy.getName();
+ }
+ String filter = properties.get("unicode_set_filter");
+ if (filter != null && filter.isEmpty()) {
+ // BE treats an explicit empty string like an absent filter.
+ properties.remove("unicode_set_filter");
+ } else if (filter != null) {
+ try {
+ UnicodeSet unicodeSet = new UnicodeSet(filter);
+ if (unicodeSet.isEmpty()) {
+ properties.remove("unicode_set_filter");
+ } else {
+ properties.put("unicode_set_filter",
unicodeSet.toPattern(false));
+ }
+ } catch (IllegalArgumentException e) {
+ // Invalid policies keep their original identity.
}
+ }
+ if (hasMode) {
+ canonicalizeIcuNormalizerMode(properties);
+ }
+ }
- Map<String, String> props = policy.getProperties();
- if (props == null || props.isEmpty()) {
- return name;
+ private static void canonicalizeIcuNormalizerMode(TreeMap<String, String>
properties) {
+ removeStringDefault(properties, "mode", "compose");
+ if (!"decompose".equals(properties.get("mode"))) {
+ return;
+ }
+ // BE ignores mode for nfd/nfkd, and nfc/nfkc in decompose mode are
the same ICU instances.
+ String name = properties.get("name");
+ if ("nfc".equals(name) || "nfd".equals(name)) {
+ properties.put("name", "nfd");
+ properties.remove("mode");
+ } else if ("nfkc".equals(name) || "nfkd".equals(name)) {
+ properties.put("name", "nfkd");
+ properties.remove("mode");
+ }
+ }
+
+ // BE reads these settings as unordered sets of trimmed, non-empty words.
+ private static void canonicalizeWordSet(TreeMap<String, String>
properties, String key) {
+ String value = properties.get(key);
+ if (value == null) {
+ return;
+ }
+ TreeSet<String> words = new TreeSet<>();
+ for (String word : value.split(",")) {
+ String trimmed = trimAsciiWhitespace(word);
+ if (!trimmed.isEmpty()) {
+ words.add(trimmed);
}
+ }
+ if (words.isEmpty()) {
+ properties.remove(key);
+ } else {
+ properties.put(key, String.join(",", words));
+ }
+ }
+
+ // BE matches custom token characters as a code point set.
+ private static void canonicalizeCustomTokenChars(TreeMap<String, String>
properties) {
+ String value = properties.get("custom_token_chars");
+ if (value == null) {
+ return;
+ }
+ StringBuilder canonical = new StringBuilder();
+
value.codePoints().distinct().sorted().forEach(canonical::appendCodePoint);
+ properties.put("custom_token_chars", canonical.toString());
+ }
- // Build identity from sorted properties
- TreeMap<String, String> sortedProps = new TreeMap<>(props);
- if (expectedType == IndexPolicyTypeEnum.TOKENIZER
- && "ngram".equals(sortedProps.get(IndexPolicy.PROP_TYPE)))
{
- // This setting only limits policy creation; it does not
change emitted tokens.
- sortedProps.remove(PROP_MAX_NGRAM_DIFF);
+ // BE collects tokenize_on_chars entries into sets, so order and repeats
do not matter.
+ private static void canonicalizeTokenizeOnChars(TreeMap<String, String>
properties) {
+ List<String> entries =
parseEntryList(properties.get("tokenize_on_chars"));
+ if (entries == null) {
+ return;
+ }
+ putEntryList(properties, "tokenize_on_chars", new TreeSet<>(entries));
+ }
+
+ // BE builds a per-character type map where a later rule for the same
character wins.
+ private static void canonicalizeTypeTable(TreeMap<String, String>
properties) {
+ List<String> rules = parseEntryList(properties.get("type_table"));
+ if (rules == null) {
+ return;
+ }
+ TreeMap<Integer, String> types = new TreeMap<>();
+ for (String rule : rules) {
+ int arrow = rule.lastIndexOf("=>");
+ if (arrow < 0 || rule.indexOf('\n') >= 0 || rule.indexOf('\r') >=
0) {
+ return;
}
- return sortedProps.toString();
- } catch (RuntimeException e) {
- return name;
+ String character = trimAsciiWhitespace(rule.substring(0, arrow));
+ String type = trimAsciiWhitespace(rule.substring(arrow + 2));
+ // Escaped characters keep the original identity rather than
reproducing BE unescaping.
+ if (character.indexOf('\\') >= 0 || character.codePointCount(0,
character.length()) != 1
+ || !WORD_DELIMITER_TYPES.contains(type)) {
+ return;
+ }
+ types.put(character.codePointAt(0), type);
+ }
+ List<String> canonicalRules = new ArrayList<>();
+ for (Map.Entry<Integer, String> entry : types.entrySet()) {
+ canonicalRules.add(new String(Character.toChars(entry.getKey())) +
"=>" + entry.getValue());
+ }
+ putEntryList(properties, "type_table", canonicalRules);
+ }
+
+ /** Parse a bracketed entry list as BE does, or return null for a
malformed list. */
+ private static List<String> parseEntryList(String value) {
+ if (value == null) {
+ return null;
+ }
+ List<String> entries = new ArrayList<>();
+ String trimmed = trimAsciiWhitespace(value);
+ if (trimmed.isEmpty()) {
+ return entries;
+ }
+ for (String item : ENTRY_SEPARATOR.split(trimmed)) {
+ String entry = trimAsciiWhitespace(item);
+ if (entry.length() < 2 || entry.charAt(0) != '[' ||
entry.charAt(entry.length() - 1) != ']') {
+ return null;
+ }
+ String content = entry.substring(1, entry.length() - 1);
+ if (!content.isEmpty()) {
+ entries.add(content);
+ }
+ }
+ return entries;
+ }
+
+ private static void putEntryList(TreeMap<String, String> properties,
String key, Collection<String> entries) {
+ if (entries.isEmpty()) {
+ properties.remove(key);
+ return;
+ }
+ StringBuilder canonical = new StringBuilder();
+ for (String entry : entries) {
+ if (canonical.length() > 0) {
+ canonical.append(",");
+ }
+ canonical.append("[").append(entry).append("]");
+ }
+ properties.put(key, canonical.toString());
+ }
+
+ // Trim the same ASCII whitespace that BE trims.
+ private static String trimAsciiWhitespace(String value) {
+ int begin = 0;
+ int end = value.length();
+ while (begin < end && isAsciiWhitespace(value.charAt(begin))) {
+ ++begin;
+ }
+ while (end > begin && isAsciiWhitespace(value.charAt(end - 1))) {
+ --end;
+ }
+ return value.substring(begin, end);
+ }
+
+ private static boolean isAsciiWhitespace(char value) {
+ return value == ' ' || (value >= '\t' && value <= '\r');
+ }
+
+ private static void canonicalizeBasicExtraChars(TreeMap<String, String>
properties) {
Review Comment:
[P1] Remove ASCII alphanumerics from Basic's effective `extra_chars`
identity. `BasicTokenizer::cut()` consumes an alphanumeric run before it ever
consults `_extra_char_set`, so `{type=basic}` and `{type=basic,extra_chars=A0}`
emit identical terms, positions, offsets, and provenance for every input.
Retaining `A0` here gives differently named aliases different identities and
lets both through CREATE/ALTER duplicate checks. This is residual to the
earlier sort/deduplicate fix because these set members are unreachable. Drop
`[A-Za-z0-9]`, remove the property when empty, and cover an alphanumeric-only
value plus a punctuation negative.
##########
fe/fe-core/src/main/java/org/apache/doris/analysis/InvertedIndexUtil.java:
##########
@@ -393,16 +420,40 @@ public static boolean
canHaveMultipleInvertedIndexes(DataType colType, List<Inde
}
Set<String> analyzerKeys = new HashSet<>();
+ Set<String> analyzerSelectors = new HashSet<>();
for (IndexDefinition indexDef : indexDefs) {
- String key = buildAnalyzerIdentity(indexDef.getProperties());
+ Map<String, String> properties = indexDef.getProperties();
+ String key = buildAnalyzerIdentity(properties);
Review Comment:
[P1] Make this identity fence equate the built-in `lowercase` normalizer
with its actual BE pipeline. FE currently produces `normalizer:lowercase` for
the built-in but `NORMALIZER:token_filter=lowercase;` for a custom normalizer
using that filter. BE builds both as the same keyword-plus-lowercase
`CustomNormalizer`, so they emit identical terms, offsets, and provenance;
their different names also evade the selector fence, letting both through
CREATE/ALTER. Serialize the built-in as the effective custom pipeline
(including consistent outer-fold context) and cover built-in-versus-custom
aliases in both DDL paths.
##########
fe/fe-core/src/main/java/org/apache/doris/analysis/invertedindex/AnalyzerIdentityBuilder.java:
##########
@@ -150,49 +221,458 @@ private static String
buildIdentityFromPolicyProperties(IndexPolicyTypeEnum type
* Resolve a component (tokenizer) to its identity.
*/
private static String resolveComponentIdentity(String name,
IndexPolicyTypeEnum expectedType) {
+ return resolveComponentIdentity(name, expectedType, null);
+ }
+
+ /**
+ * {@code foldBlockedBytes} is the case-folding context of a char filter:
null without a
+ * downstream fold, otherwise the bytes that filters between this one and
the fold rewrite.
+ */
+ private static String resolveComponentIdentity(
+ String name, IndexPolicyTypeEnum expectedType, boolean[]
foldBlockedBytes) {
if (Strings.isNullOrEmpty(name)) {
return "";
}
- // Check if it's a built-in component
- if (expectedType == IndexPolicyTypeEnum.TOKENIZER
- && IndexPolicy.BUILTIN_TOKENIZERS.contains(name)) {
- return name;
+ // Existing named policies take precedence over built-ins for upgrade
compatibility.
+ try {
+ Env env = Env.getCurrentEnv();
+ if (env != null && env.getIndexPolicyMgr() != null) {
+ IndexPolicy policy =
env.getIndexPolicyMgr().getPolicyByName(name);
+ if (policy != null && policy.getType() == expectedType) {
+ if (policy.isInvalid()) {
+ return "invalid-policy:" + policy.getId() + ":" +
policy.getName();
+ }
+ Map<String, String> props = policy.getProperties();
+ if (props != null && !props.isEmpty()) {
+ TreeMap<String, String> sortedProps = new
TreeMap<>(props);
+ String type = sortedProps.get(IndexPolicy.PROP_TYPE);
+ String normalizedType =
normalizeBuiltinComponentName(type, expectedType);
+ if (normalizedType != null) {
+ if ("empty".equals(normalizedType)) {
+ return "";
+ }
+ sortedProps.put(IndexPolicy.PROP_TYPE,
normalizedType);
+ canonicalizeEffectiveComponentProperties(
+ sortedProps, normalizedType, expectedType);
+ if (sortedProps.size() == 1) {
+ return normalizedType;
+ }
+ }
+ if (expectedType == IndexPolicyTypeEnum.TOKENIZER
+ &&
"ngram".equals(sortedProps.get(IndexPolicy.PROP_TYPE))) {
+ // This setting only limits policy creation; it
does not change emitted tokens.
+ sortedProps.remove(PROP_MAX_NGRAM_DIFF);
+ }
+ if (expectedType == IndexPolicyTypeEnum.CHAR_FILTER
+ &&
"char_replace".equals(sortedProps.get(IndexPolicy.PROP_TYPE))) {
+ String replacement =
sortedProps.getOrDefault("replacement", " ");
+ String pattern = canonicalizeCharReplacePattern(
+ sortedProps.get("pattern"), replacement,
foldBlockedBytes);
+ if (pattern.isEmpty()) {
+ return "";
+ }
+ sortedProps.put("pattern", pattern);
+ sortedProps.put("replacement", replacement);
+ }
+ if (normalizedType != null && sortedProps.size() == 1)
{
+ return normalizedType;
+ }
+ return sortedProps.toString();
+ }
+ }
+ }
+ } catch (RuntimeException e) {
+ // Fall through to built-in resolution or the original name.
+ }
+
+ String normalizedName = normalizeBuiltinComponentName(name,
expectedType);
+ return "empty".equals(normalizedName) ? "" : normalizedName == null ?
name : normalizedName;
+ }
+
+ private static void canonicalizeEffectiveComponentProperties(
+ TreeMap<String, String> properties, String type,
IndexPolicyTypeEnum expectedType) {
+ if ("pinyin".equals(type)) {
+ removeBooleanDefaults(properties, true,
+ "keep_first_letter", "keep_full_pinyin",
"keep_none_chinese",
+ "keep_none_chinese_together",
"keep_none_chinese_in_first_letter",
+ "lowercase", "trim_whitespace", "ignore_pinyin_offset",
+ "none_chinese_pinyin_tokenize");
+ removeBooleanDefaults(properties, false,
+ "keep_separate_first_letter", "keep_joined_full_pinyin",
"keep_original",
+ "keep_none_chinese_in_joined_full_pinyin",
"remove_duplicated_term",
+ "fixed_pinyin_offset", "keep_separate_chinese");
+ removeIntegerDefault(properties, "limit_first_letter_length", 16);
+ canonicalizePinyinDependencies(properties, expectedType);
+ return;
+ }
+
+ if (expectedType == IndexPolicyTypeEnum.TOKEN_FILTER) {
+ if ("asciifolding".equals(type)) {
+ removeBooleanDefaults(properties, false, "preserve_original");
+ } else if ("word_delimiter".equals(type)) {
+ removeBooleanDefaults(properties, true, "generate_word_parts",
"generate_number_parts",
+ "split_on_case_change", "split_on_numerics",
"stem_english_possessive");
+ removeBooleanDefaults(properties, false, "catenate_words",
"catenate_numbers",
+ "catenate_all", "preserve_original");
+ canonicalizeWordSet(properties, "protected_words");
+ canonicalizeTypeTable(properties);
+ } else if ("icu_normalizer".equals(type)) {
+ canonicalizeIcuNormalizerDefaults(properties, false);
+ }
+ return;
}
- // For custom component, get its properties
+ if (expectedType == IndexPolicyTypeEnum.CHAR_FILTER) {
+ if ("icu_normalizer".equals(type)) {
+ canonicalizeIcuNormalizerDefaults(properties, true);
+ }
+ return;
+ }
+
+ if (expectedType != IndexPolicyTypeEnum.TOKENIZER) {
+ return;
+ }
+ switch (type) {
+ case "ngram":
+ case "edge_ngram":
+ removeIntegerDefault(properties, "min_gram", 1);
+ removeIntegerDefault(properties, "max_gram", 2);
+ canonicalizeWordSet(properties, "token_chars");
Review Comment:
[P1] Canonicalize the union of named and custom NGram matchers, not the two
collections independently. BE ORs every matcher in `CompositeMatcher`; with
`letter` selected, custom code point `A` adds nothing. Thus
`{token_chars=letter}` and `{token_chars=letter,custom; custom_token_chars=A}`
behave identically for NGram and EdgeNGram, yet receive different identities
and can pass CREATE/ALTER as aliases. This is residual to collection
ordering/deduplication. Remove custom members already matched by named classes,
then drop an empty `custom` branch, with an uncovered punctuation point as a
negative.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
