mistocr 0.1.1__py3-none-any.whl → 0.1.2__py3-none-any.whl

mistocr/__init__.py

@@ -1 +1 @@
-__version__ = "0.1.1"
+__version__ = "0.1.2"

mistocr/refine.py

@@ -43,27 +43,30 @@ prompt_fix_hdgs = """Fix markdown heading hierarchy errors while preserving the
 
 INPUT FORMAT: Each heading is prefixed with its index number (e.g., "0. # Title")
 
-RULES - Only fix these errors:
-1. **Level jumps**: Headings can only increase by one # at a time
-   - Wrong: 0. # Title → 1. #### Abstract
-   - Fixed: 0. # Title → 1. ## Abstract
+RULES - Apply these fixes in order:
 
-2. **Numbering inconsistency**: Subsection numbers must be one level deeper
-   - Wrong: 4. ## 3. Section → 5. ## 3.1 Subsection
-   - Fixed: 4. ## 3. Section → 5. ### 3.1 Subsection
+1. **Single H1 rule**: Documents must have exactly ONE # heading (the title/main heading)
+   - All other headings should be ## or deeper
 
-3. **Preserve working structure**: If sections are consistently marked, keep it
+2. **Infer depth from numbering patterns**: If headings contain section numbers, deeper nesting means deeper heading level
+   - Parent section (e.g., "1", "2", "A") should be shallower than child (e.g., "1.1", "2.a", "A.1")
+   - Child section should be one # deeper than parent
+   - Works with any numbering: "1/1.1/1.1.1", "A/A.1/A.1.a", "I/I.A/I.A.1", etc.
 
-4. **Decreasing levels is OK**: Going from ### to ## is valid for new sections
+3. **Level jumps**: Headings can only increase by one # at a time when moving deeper
+   - Wrong: ## Section → ##### Subsection
+   - Fixed: ## Section → ### Subsection
+
+4. **Decreasing levels is OK**: Moving back up the hierarchy (### to ##) is valid for new sections
 
 OUTPUT: Return a Python dictionary mapping index to corrected heading (without the index prefix).
-Only include entries that need changes. Example: {{1: '## Abstract', 15: '### PASCAL VOC'}}
+Only include entries that need changes.
 
 Headings to analyze:
 {headings_list}
 """
 
-# %% ../nbs/01_refine.ipynb 18
+# %% ../nbs/01_refine.ipynb 16
 def fix_hdg_hierarchy(
     hdgs: list[str], # List of markdown headings
     prompt: str=prompt_fix_hdgs, # Prompt to use
@@ -79,7 +82,7 @@ def fix_hdg_hierarchy(
         )
     return json.loads(r.choices[0].message.content)['corrections']
 
-# %% ../nbs/01_refine.ipynb 21
+# %% ../nbs/01_refine.ipynb 19
 def mk_fixes_lut(
     hdgs: list[str], # List of markdown headings
     model: str='claude-sonnet-4-5', # Model to use
@@ -89,7 +92,7 @@ def mk_fixes_lut(
     fixes = fix_hdg_hierarchy(hdgs, model, api_key)
     return {hdgs[int(k)]:v for k,v in fixes.items()}
 
-# %% ../nbs/01_refine.ipynb 24
+# %% ../nbs/01_refine.ipynb 22
 def apply_hdg_fixes(
     p:str, # Page to fix
     lut_fixes: dict[str, str], # Lookup table of fixes
@@ -99,7 +102,7 @@ def apply_hdg_fixes(
     for old in get_hdgs(p): p = p.replace(old, lut_fixes.get(old, old) + (f' .... page {pg}' if pg else ''))
     return p
 
-# %% ../nbs/01_refine.ipynb 27
+# %% ../nbs/01_refine.ipynb 25
 def fix_md_hdgs(
     src:str, # Source directory with markdown pages
     model:str='claude-sonnet-4-5', # Model

{mistocr-0.1.1.dist-info → mistocr-0.1.2.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: mistocr
-Version: 0.1.1
+Version: 0.1.2
 Summary: Simple batch OCR for PDFs using Mistral's state-of-the-art vision model
 Home-page: https://github.com/franckalbinet/mistocr
 Author: Solveit

mistocr-0.1.2.dist-info/RECORD

@@ -0,0 +1,10 @@
+mistocr/__init__.py,sha256=YvuYzWnKtqBb-IqG8HAu-nhIYAsgj9Vmc_b9o7vO-js,22
+mistocr/_modidx.py,sha256=R9zVMv4dKz2sLStoB5wBoKRqjza216_z8xPXszoplU4,2660
+mistocr/core.py,sha256=wtaYZ_Fz0dXIb1cYLAtymwR9Z7_KBI4ULy-UpM3YTX4,7154
+mistocr/refine.py,sha256=kZH-z55zNcIgmHSpHETkHjGCoZENmMvL9QJOpNic2GI,4531
+mistocr-0.1.2.dist-info/licenses/LICENSE,sha256=xV8xoN4VOL0uw9X8RSs2IMuD_Ss_a9yAbtGNeBWZwnw,11337
+mistocr-0.1.2.dist-info/METADATA,sha256=AOoj8PRpmJDlfRgEFzfR7UQYa7ZI5cVTsXlopHzk6JM,4848
+mistocr-0.1.2.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+mistocr-0.1.2.dist-info/entry_points.txt,sha256=HjAiHozobM-alm_6bTF-ehRr2DD3KYE9PgRngelONOY,36
+mistocr-0.1.2.dist-info/top_level.txt,sha256=LelTYnSpSXLy1Hb1m2YP3gt8luwP-I8KV0NjP_ucdSs,8
+mistocr-0.1.2.dist-info/RECORD,,

mistocr-0.1.1.dist-info/RECORD

@@ -1,10 +0,0 @@
-mistocr/__init__.py,sha256=rnObPjuBcEStqSO0S6gsdS_ot8ITOQjVj_-P1LUUYpg,22
-mistocr/_modidx.py,sha256=R9zVMv4dKz2sLStoB5wBoKRqjza216_z8xPXszoplU4,2660
-mistocr/core.py,sha256=wtaYZ_Fz0dXIb1cYLAtymwR9Z7_KBI4ULy-UpM3YTX4,7154
-mistocr/refine.py,sha256=0N0omMZvPqydArPaiTdkX8tts4eS4AgfAJQP98WnvwY,4307
-mistocr-0.1.1.dist-info/licenses/LICENSE,sha256=xV8xoN4VOL0uw9X8RSs2IMuD_Ss_a9yAbtGNeBWZwnw,11337
-mistocr-0.1.1.dist-info/METADATA,sha256=ZPvywkk_QS7UTfPtup5IppiLzf5aZ1tCSPkZLV9g4WE,4848
-mistocr-0.1.1.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-mistocr-0.1.1.dist-info/entry_points.txt,sha256=HjAiHozobM-alm_6bTF-ehRr2DD3KYE9PgRngelONOY,36
-mistocr-0.1.1.dist-info/top_level.txt,sha256=LelTYnSpSXLy1Hb1m2YP3gt8luwP-I8KV0NjP_ucdSs,8
-mistocr-0.1.1.dist-info/RECORD,,