129 Commits

Author SHA1 Message Date
Zhichang Yu
0151d42156
Reuse loaded modules if possible (#5231)
### What problem does this PR solve?

Reuse loaded modules if possible

### Type of change

- [x] Refactoring
2025-02-21 17:21:01 +08:00
Zhichang Yu
c326f14fed
Optimized Recognizer.sort_X_firstly and Recognizer.sort_Y_firstly (#5182)
### What problem does this PR solve?

Optimized Recognizer.sort_X_firstly and Recognizer.sort_Y_firstly

### Type of change

- [x] Performance Improvement
2025-02-20 15:41:12 +08:00
Kevin Hu
b08bb56f6c
Display thinking for deepseek r1 (#4904)
### What problem does this PR solve?
#4903
### Type of change

- [x] New Feature (non-breaking change which adds functionality)
2025-02-12 15:43:13 +08:00
Mathias Panzenböck
6b389e01b5
Remove use of eval() from operators.py (#4888)
Use `np.float32()` instead.

### What problem does this PR solve?

Using `eval()` can lead to code injections.

I think `eval()` is only used to parse a floating point number here.
This change preserves the correct behavior if the string `"None"` is
supplied. But if that behavior isn't intended then this part could be
just deleted instead, since `np.float32()` is parsing strings anyway:

```Python
        if isinstance(scale, str):
            scale = eval(scale)
```

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
2025-02-12 12:53:42 +08:00
SkyfireWXY
8fcca1b958
fix: big xls file error (#4859)
### What problem does this PR solve?

if *.xls file is too large, .eg >50M, I get error.

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
2025-02-12 12:39:25 +08:00
Zhichang Yu
3411d0a2ce
Added cuda_is_available (#4725)
### What problem does this PR solve?

Added cuda_is_available

### Type of change

- [x] Refactoring
2025-02-05 18:01:23 +08:00
Zhichang Yu
e1526846da
Fixed GPU detection on CPU only environment (#4711)
### What problem does this PR solve?

Fixed GPU detection on CPU only environment. Close #4692

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
2025-02-05 12:02:43 +08:00
Kevin Hu
6f30397bb5
Infinity adapt to graphrag. (#4663)
### What problem does this PR solve?


### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
2025-01-27 18:35:18 +08:00
Kevin Hu
1bff6b7333
Fix t_ocr.py for PNG image. (#4625)
### What problem does this PR solve?
#4586

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
2025-01-24 11:47:27 +08:00
Zhichang Yu
4230402fbb
deepdoc use GPU if possible (#4618)
### What problem does this PR solve?

deepdoc use GPU if possible

### Type of change

- [x] Refactoring
2025-01-24 09:48:02 +08:00
Mathias Panzenböck
1a367664f1
Remove usage of eval() from postprocess.py (#4571)
Remove usage of `eval()` from postprocess.py

### What problem does this PR solve?

The use of `eval()` is a potential security risk. While the use of
`eval()` is guarded and thus not a security risk normally, `assert`s
aren't run if `-O` or `-OO` is passed to the interpreter, and as such
then the guard would not apply. In any case there is no reason to use
`eval()` here at all.

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
- [x] Other (please describe):

Potential security fix if somehow the passed `modul_name` could be user
controlled.
2025-01-22 19:37:24 +08:00
Jin Hai
3894de895b
Update comments (#4569)
### What problem does this PR solve?

Add license statement.

### Type of change

- [x] Refactoring

Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2025-01-21 20:52:28 +08:00
Mathias Panzenböck
75e1981e13
Remove use of eval() from recognizer.py (#4480)
`eval(op_type)` -> `getattr(operators, op_type)`

### What problem does this PR solve?

Using `eval()` can lead to code injections and is entirely unnecessary
here.

### Type of change

- [x] Other (please describe):

Best practice code improvement, preventing the possibility of code
injection.
2025-01-20 09:52:47 +08:00
Mathias Panzenböck
4f9f9405b8
Remove use of eval() from ocr.py (#4481)
`eval(op_name)` -> `getattr(operators, op_name)`

### What problem does this PR solve?

Using `eval()` can lead to code injections and is entirely unnecessary
here.

### Type of change

- [x] Other (please describe):

Best practice code improvement, preventing the possibility of code
injection.
2025-01-20 09:52:30 +08:00
Kevin Hu
c852a6dfbf
Accelerate titles' embeddings. (#4492)
### What problem does this PR solve?


### Type of change

- [x] Performance Improvement
2025-01-15 15:20:29 +08:00
Kevin Hu
e478586a8e
Refactor. (#4487)
### What problem does this PR solve?

### Type of change

- [x] Refactoring
2025-01-15 14:06:46 +08:00
Zhi-Qiang You
b7ce4e7e62
fix:t_recognizer TypeError: 'super' object is not callable (#4404)
### What problem does this PR solve?

[Bug]: layout recognizer failed for wrong boxes class type #4230
(https://github.com/infiniflow/ragflow/issues/4230)

### Type of change

- [ ] Bug Fix (non-breaking change which fixes an issue)

---------

Co-authored-by: youzhiqiang <zhiqiang.you@aminer.com>
Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
2025-01-08 10:59:35 +08:00
Kevin Hu
2e40c2a6f6
Fix t_recognizer issue. (#4387)
### What problem does this PR solve?

#4230

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
2025-01-07 13:17:46 +08:00
Kevin Hu
983ec0666c
Fix param error. (#4355)
### What problem does this PR solve?

#4230

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
2025-01-06 13:54:17 +08:00
Kevin Hu
59a78408be
Fix t_recognizer.py after model updating. (#4330)
### What problem does this PR solve?

#4230

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
2025-01-02 17:00:11 +08:00
Kevin Hu
76cd23eecf
Catch the exception while parsing pptx. (#4202)
### What problem does this PR solve?
#4189

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
2024-12-24 10:49:28 +08:00
Kevin Hu
2cbe064080
Add Llama3.3 (#4174)
### What problem does this PR solve?

#4168

### Type of change

- [x] New Feature (non-breaking change which adds functionality)
2024-12-23 11:18:01 +08:00
ly0303521
101b8ff813
fix chunk method "Table" losing content when the Excel file has multi… (#4123)
…ple sheets

### What problem does this PR solve?
discussed in https://github.com/infiniflow/ragflow/pull/4102
- In excel_parser.py, `total` means the total number of rows in Excel,
but it return in the first iterate, that lead to the wrong `to_page`
- In table.py, it when Excel file has multiple sheets, it will be
divided into multiple parts, every part size is 3000, `data` may be
empty, because it has recorded in the last iterate.
### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
2024-12-19 17:30:26 +08:00
Kevin Hu
ce1e855328
Upgrades Document Layout Analysis model. (#4054)
### What problem does this PR solve?

#4052

### Type of change

- [x] New Feature (non-breaking change which adds functionality)
2024-12-17 11:27:19 +08:00
Jin Hai
275b5d14f2
Fix json file parse (#4004)
### What problem does this PR solve?

Fix json file parsing

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

---------

Signed-off-by: jinhai <haijin.chn@gmail.com>
2024-12-12 20:34:46 +08:00
Zhichang Yu
9a6d976252
Add back beartype (#3967)
### What problem does this PR solve?

Add back beartype

### Type of change

- [x] Refactoring
2024-12-10 18:43:43 +08:00
Zhichang Yu
1254ecf445
Added static check at PR CI (#3921)
### What problem does this PR solve?

Added static check at PR CI

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
- [x] Refactoring
2024-12-08 21:23:51 +08:00
Zhichang Yu
0d68a6cd1b
Fix errors detected by Ruff (#3918)
### What problem does this PR solve?

Fix errors detected by Ruff

### Type of change

- [x] Refactoring
2024-12-08 14:21:12 +08:00
Jin Hai
821fdf02b4
Fix parsing JSON file error (#3829)
### What problem does this PR solve?

Close issue: #3828

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

Signed-off-by: jinhai <haijin.chn@gmail.com>
2024-12-03 19:02:03 +08:00
Yuhao Tsui
7b6a5ffaff
Fix: page_chars attribute does not exist in some formats of PDF (#3796)
### What problem does this PR solve?

In #3335 someone suggested to upgrade pdfplumber==0.11.1, but that
didn't solve it.
It's actually the special formatting in some of the pdfs that triggers
the problem.

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
2024-12-03 11:08:06 +08:00
Kevin Hu
7058ac0041
Fix out of boundary. (#3786)
### What problem does this PR solve?

#3769
### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
2024-12-02 11:38:53 +08:00
Zhichang Yu
bc701d7b4c
Edit chunk shall update instead of insert it (#3709)
### What problem does this PR solve?

Edit chunk shall update instead of insert it. Close #3679 

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
2024-11-28 13:00:38 +08:00
Zhichang Yu
2249d5d413
Always open text file for write with UTF-8 (#3688)
### What problem does this PR solve?

Always open text file for write with UTF-8. Close #932 

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
2024-11-27 16:24:16 +08:00
Zhichang Yu
cad341e794 Added kb_id filter to knn. Fix #3458 (#3513)
### What problem does this PR solve?

Added kb_id filter to knn. Fix #3458

- [x] Bug Fix (non-breaking change which fixes an issue)
2024-11-20 20:53:30 +08:00
Zhichang Yu
4413683898
Introduced beartype (#3460)
### What problem does this PR solve?

Introduced [beartype](https://github.com/beartype/beartype) for runtime
type-checking.

### Type of change

- [x] New Feature (non-breaking change which adds functionality)
2024-11-18 17:38:17 +08:00
Jin Hai
1e90a1bf36
Move settings initialization after module init phase (#3438)
### What problem does this PR solve?

1. Module init won't connect database any more.
2. Config in settings need to be used with settings.CONFIG_NAME

### Type of change

- [x] Refactoring

Signed-off-by: jinhai <haijin.chn@gmail.com>
2024-11-15 17:30:56 +08:00
Zhichang Yu
30f6421760
Use consistent log file names, introduced initLogger (#3403)
### What problem does this PR solve?

Use consistent log file names, introduced initLogger

### Type of change

- [ ] Bug Fix (non-breaking change which fixes an issue)
- [ ] New Feature (non-breaking change which adds functionality)
- [ ] Documentation Update
- [x] Refactoring
- [ ] Performance Improvement
- [ ] Other (please describe):
2024-11-14 17:13:48 +08:00
Kevin Hu
4caf932808
fix bug about fetching knowledge graph (#3394)
### What problem does this PR solve?


### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
2024-11-14 12:29:15 +08:00
Zhichang Yu
a2a5631da4
Rework logging (#3358)
Unified all log files into one.

### What problem does this PR solve?

Unified all log files into one.

### Type of change

- [x] Refactoring
2024-11-12 17:35:13 +08:00
kuschzzp
9c6cc20356
Fix:#3230 When parsing a docx file using the Book parsing method, to_page is always -1, resulting in a block count of 0 even if parsing is successful (#3249)
### What problem does this PR solve?

When parsing a docx file using the Book parsing method, to_page is
always -1, resulting in a block count of 0 even if parsing is successful

Fix:#3230

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
2024-11-08 09:21:42 +08:00
Kevin Hu
2d1fbefdb5
search between multiple indiices for team function (#3079)
### What problem does this PR solve?

#2834 
### Type of change

- [x] New Feature (non-breaking change which adds functionality)
2024-10-29 13:19:01 +08:00
Kevin Hu
bfc07fe4f9
bigger resolution for OCR (#2919)
### What problem does this PR solve?



### Type of change

- [x] Performance Improvement
2024-10-21 16:25:42 +08:00
chongchuanbing
66172cef3e
fix: torch dependency start error (#2777)
### What problem does this PR solve?

when use slim image, remove ```torch``` denpendency.

### Type of change

- [✓] Bug Fix (non-breaking change which fixes an issue)

---------

Co-authored-by: chongchuanbing <chongchuanbing@gmail.com>
Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
2024-10-10 10:06:03 +08:00
Ikko Eltociear Ashimine
c552a02e7f
chore: update operators.py (#2724)
### What problem does this PR solve?
substract -> subtract
### Type of change

- [ ] Bug Fix (non-breaking change which fixes an issue)
- [ ] New Feature (non-breaking change which adds functionality)
- [ ] Documentation Update
- [x] Refactoring
- [ ] Performance Improvement
- [ ] Other (please describe):
2024-10-08 10:34:52 +08:00
Kevin Hu
daa65199e8
trival (#2650)
### What problem does this PR solve?

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
2024-09-29 13:20:02 +08:00
Kevin Hu
fc867cb959
rename get_txt to get_text (#2649)
### What problem does this PR solve?



### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
2024-09-29 12:47:09 +08:00
yqkcn
aea553c3a8
Add get_txt function (#2639)
### What problem does this PR solve?

Add get_txt function to reduce duplicate code

### Type of change

- [x] Refactoring

---------

Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
2024-09-29 10:29:56 +08:00
liuhua
b68d349bd6
Fix: renrank_model and pdf_parser bugs | Update: session API (#2601)
### What problem does this PR solve?

Fix: renrank_model and pdf_parser bugs | Update: session API
#2575
#2559
### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
- [x] Refactoring

---------

Co-authored-by: liuhua <10215101452@stu.ecun.edu.cn>
2024-09-26 16:05:25 +08:00
Kevin Hu
7bb28ca2bd
add lighten control (#2567)
### What problem does this PR solve?

#2295

### Type of change

- [ ] Bug Fix (non-breaking change which fixes an issue)
- [x] New Feature (non-breaking change which adds functionality)
- [ ] Documentation Update
- [ ] Refactoring
- [ ] Performance Improvement
- [ ] Other (please describe):
2024-09-24 19:22:01 +08:00
Vitaliy Groshev
7e75b9d778
fix parsing spaces in russian language PDFs (#1987) (#2427)
### What problem does this PR solve?

[#1987](https://github.com/infiniflow/ragflow/issues/1987)

When scanning PDF files character by character, the parser excluded
spaces if the string did not match regex. Text from [Russian
documents](https://github.com/user-attachments/files/16659706/dogovor_oferta.pdf)
needs spaces, but it does not match the regex because it uses different
alphabet. That's why PDFs were parsed incorrectly and were almost
unusable as source. Fixed that by adding Russian alphabet to regex.

There might be problems with other languages that use different
alphabets. I additionally tested [PDF in
Spanish](https://www.scusd.edu/sites/main/files/file-attachments/howtohelpyourchildsucceedinschoolspanish.pdf?1338307816)
and old [a-zA-Z...] regex parses it correctly with spaces.

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
2024-09-14 13:14:39 +08:00