feat(sql): implement LOAD_FILE and AI_SPLIT_DOCUMENT

[7774596]

Task Description

Add LOAD_FILE(location, file) to read external files via LOCATION and return BLOB. Add AI_SPLIT_DOCUMENT(content[, params_json]) as a table function that outputs chunk_id, chunk_offset, chunk_length, and chunk_text rows with text/markdown splitting and sentence or word modes.

Passed Regressions

• tools/deploy/mysql_test/test_suite/ai_funcs/t/load_file.test
• tools/deploy/mysql_test/test_suite/ai_funcs/t/ai_split_document.test

TeamID

Dase314

[CLAassistant]

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you sign our Contributor License Agreement before we can accept your contribution.
_{You have signed the CLA already but the status is still pending? Let us recheck it.}

[LINxiansheng]

Document AI & IK Custom Dictionary Score

Document AI Functions Score
===========================
score: 100.00 / 100
load_file: 50 / 50
ai_split_document: 50 / 50

IK Custom Dictionary Score
==========================
score: 0.00 / 100
ik_custom_dict: 0 / 100

FTS Large Benchmark Score

FTS Large Benchmark Score
=========================
score: 1.94 / 100
mean_improvement: 0.97%
full_score_improvement: 50.00%

build_improvement: -0.68%
  build_ik_all_sec: baseline=35.2836, current=36.358, improvement=-3.05%
  build_ik_content_sec: baseline=28.3764, current=28.276, improvement=0.35%
  build_beng_en_sec: baseline=14.7578, current=14.662, improvement=0.65%
tokenize_improvement: 2.97%
  tokenize_ik_avg_ms: baseline=0.76478, current=0.7368, improvement=3.66%
  tokenize_beng_avg_ms: baseline=0.42262, current=0.413, improvement=2.28%
query_improvement: 0.62%
  query_cn_avg_ms: baseline=16.6628, current=16.6081, improvement=0.33%
  query_beng_avg_ms: baseline=24.3042, current=24.1646, improvement=0.57%
  query_mixed_avg_ms: baseline=17.5593, current=17.4566, improvement=0.58%
  query_limit_avg_ms: baseline=16.2334, current=16.0718, improvement=1.00%

FTS Large Benchmark Report

========================================
FTS Large Benchmark Report
========================================
timestamp:              2026-07-14 06:00:50
label:                  vldb-ci-29309031392-1
git_head:               3065480
git_dirty:              0
rows:                   20000
batch:                  500
rounds:                 3000
query_rounds:           200
samples:                3
warmup:                 30
skip_load:              0
----------------------------------------
select1_avg_ms:         0.2077
select1_stdev_ms:       0.0042
raw_load_sec:           1.539
raw_load_rows_per_sec:  12995.5
build_ik_all_sec:       36.358
build_ik_content_sec:   28.276
build_beng_en_sec:      14.662
build_total_sec:        79.309
----------------------------------------
tokenize_ik_avg_ms:     0.7368
tokenize_ik_median_ms:  0.7367
tokenize_ik_stdev_ms:   0.0004
tokenize_beng_avg_ms:   0.4130
tokenize_beng_median_ms:0.4130
tokenize_beng_stdev_ms: 0.0002
----------------------------------------
query_cn_hits:          8001
query_cn_avg_ms:        16.6081
query_cn_stdev_ms:      0.0229
query_beng_hits:        11000
query_beng_avg_ms:      24.1646
query_beng_stdev_ms:    0.0288
query_mixed_hits:       7332
query_mixed_avg_ms:     17.4566
query_mixed_stdev_ms:   0.0073
query_limit_hits:       20
query_limit_avg_ms:     16.0718
query_limit_stdev_ms:   0.0116
========================================

Workflow run