boluor opened a new pull request, #3723: URL: https://github.com/apache/doris-website/pull/3723
## Summary A pass of small content fixes — broken anchors, wrong wording, copy-paste mistakes between examples, and bad SQL in code blocks. ### Auth / SHOW GRANTS - `authentication-and-authorization.md`: `GRANT ALTER_PRIV ON . TO user1@'ip1';` lacked a resource scope. Align with the surrounding `ON *.*` examples by using `ON *.*`. - Same file: wording said `SHOW ALL GRANTS` lists "all created users"; it actually lists grants. Reword to "view all users' permissions and roles" (matching how the same command is described later in the doc). ### Workload-management - `analysis-diagnosis.md`: `ORDER BY a, b, c DESC;` only binds DESC to the last column; add DESC to each. ### Import / load examples - `variant.md`: tutorial uses `test_variant`, but two `desc github_events;` invocations broke the flow. - `bigquery.md`, `redshift.md`, `snowflake.md`: LOAD examples submit a label `sales_data_2025_04_08` but the subsequent `SHOW LOAD` queries `label_sales_data_2025_04_08`. Align the SHOW LOAD query and the displayed `Label:` line to the actually-submitted label. - `mysql-load-manual.md`: parameter-table row used `LINE TERMINATED BY`; the syntax block and other rows use `LINES TERMINATED BY`. - `continuous-load-s3.md`, `continuous-load-postgresql-table.md`: anchor `insert-into-manual.md#load-configuration-parameters` does not exist; the target heading slugs to `#import-configuration-parameters`. - `group-commit-manual.md`: two examples used non-standard ports (`9087`, `9038`); align to the standard query port `9030`. ### `transaction.md` - `INSERT ... WHERE id = 7` reported `Query OK, 0 rows affected`, but the SELECT immediately below shows id=7 was written. Should be `1 row affected`. ### Kubernetes deploy docs - `cluster-operation.md`: under the **Upgrade FE** section, step 2 read "apply this modification to upgrade BE" — BE→FE. ### LLM / Query cache - `llm-sql-functions.mdx`: prose said "Eleven scalar functions"; the bulleted list actually has ten `AI_*` items. - `query-cache.mdx`: "four overlapping days (2024-01-05 through 2024-01-07)" — that range is three days; both the SQL comment and the prose corrected to "three". ### OS preparation - `os-checking.md`: the `vm.max_map_count` section attributed `Too many open files` to a VMA limit. That's a file-descriptor symptom; the actual VMA-limit symptom is `mmap` / `Cannot allocate memory`. ### Lakehouse best-practices - `doris-aws-s3tables.md`: `SWITCH iceberg_s3;` — that catalog was never created; the catalogs in this doc are `aws_s3_tables` and `glue_s3`. - `doris-iceberg.md`: `CREATE TABLE iceberg.nyc.taxis2 AS SELECT * FROM iceberg.nyc.taxis;` reported `6 rows affected`, but the source `taxis` table at that point has 4 rows. - `doris-onelake.md`: quoting on `iceberg.rest.oauth2.credential` value was malformed (`'<oauth2.client_id>:'<oauth2.client_secret>,`); restored to `'<oauth2.client_id>:<oauth2.client_secret>',`. ### Observability - `log.md`: Logstash `http_hosts` third element had a swapped closing quote (`"<http://fehost3:http_port";>`). - `log.md`: routine-load `COLUMNS(ts, clientip, request, status, size)` did not match the `log_table` schema, which is `ts, host, path, message`. ### Async materialized view - `faq.md`: CREATE uses `STARTS '2024-12-01...'` but the mv_infos result shows `STARTS "2025-12-01..."`. Align result to CREATE input. ### Optimization / runtime filter - `runtime-filter.md`: profile sample showed `RuntimeFilterState = false`; valid values are `READY` / `NOT_READY`. The surrounding context describes a runtime filter that did not take effect, so use `NOT_READY`. ### SQL data types and operators - `string-literal.md`: NUL entry's "meaning" cell had mojibake (`('X'00')`); replace with `(0x00)`. Also close an unclosed backtick in the notes (``and `'y' are equivalent``). - `pattern-matching-operators.md`: LIKE / REGEXP descriptions read "If `<char1>` does not match the pattern, it is TRUE" — inverted. Reworded so TRUE means matches, with `NOT` flipping. - `QUANTILE-STATE.md`: a sentence was copy-pasted from the HLL doc (references "HLL type" / "HLL_UNION"); replaced with QUANTILE_STATE / QUANTILE_UNION. - `datetime-conversion.md`: "Results are shown using DATETIME(3)" did not match the surrounding table and result rows, which all use DATETIME(6). - `cast-to-string.md`: comment "e < 7, uses scientific notation" contradicts the rule above (decimal notation when e < 7). - `GEO.md`: several CREATE TABLE statements were missing closing parens, and ten used the full-width `）` instead of `)`. ### Array / JSON functions - `array-distinct.md`: empty-array and NULL-array result blocks were missing their preceding `SELECT array_distinct(...)` statements. - `array-apply.md`: `op` was called a "binary operator" — confusing since `op` is a comparison operator. Reword to "comparison operator". - `coalesce.md`, `greatest.md`, `least.md`: the `[Type Conversion](../../../basic-element/...)` relative link was one `../` short — these files live 4 levels deep under `sql-manual/`, so prepend an extra `../`. - `period-diff.md`: description said the function returns the "absolute" difference; it returns a signed value (and the examples show negative results). - `json-parse.md`: example for `json_parse_error_to_null` passed two arguments; the function takes one (the result-table header already showed one arg). ## Scope Applied across `docs/`, `versioned_docs/version-{2.1,3.x,4.x}/`, and `i18n/zh-CN/` where the same content exists. For a few findings whose zh prose differs from the EN sentence (`QUANTILE-STATE` wording, `period-diff` "absolute", `array-apply` "binary operator", `cast-to-string` notation comment, `datetime-conversion` DATETIME(3)→(6), `os-checking` VMA paragraph), only the EN files were touched; zh equivalents would need separately translated edits. ## Test plan - [x] Spot-check each edit's diff per file. - [x] Verify each replacement anchor / catalog / function-arity / column-name matches surrounding context in the same file. - [ ] CI build (docusaurus + sidebar checks). -- 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]
