v1.2 - a better uploading feature with file merging

This commit is contained in:
JoelShepard
2025-06-10 15:01:21 +02:00
parent cfea84ca25
commit caacaa5dd5
3 changed files with 122 additions and 4 deletions

1
.gitignore vendored
View File

@@ -1,6 +1,7 @@
kindle_books/*
__pycache__/
.ropeproject
.claude/
# Configuration files with sensitive data
config.json

View File

@@ -93,3 +93,13 @@ Each book generates a Markdown file with citations formatted as blockquotes:
## License
This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.
## v1.2
Novità in questa release:
- Merge avanzato di citazioni locali e remote con risoluzione carattere-per-carattere
- Risoluzione dei conflitti: si mantiene la versione remota per modifiche minori e quella locale per modifiche maggiori
- Preservazione delle righe vuote e rimozione automatica di backslash indesiderati attorno a `[[…]]`
- Gestione dei nomi file URL-encoded per corretto matching
- Nuovo flag CLI `--clear` per eliminare esplicitamente i file remoti prima dell'upload

115
main.py
View File

@@ -491,6 +491,113 @@ def upload_files_to_webdav(file_paths):
print("Error: WebDAV username or password not configured")
return [], file_paths
auth = HTTPBasicAuth(WEBDAV_USERNAME, WEBDAV_PASSWORD)
# Merge local and remote changes before upload
print("Merging local and remote changes...")
from urllib.parse import urlparse
from pathlib import Path
# Retrieve list of remote files
remote_urls = get_webdav_directory_contents(WEBDAV_BASE_URL, auth)
parsed_base = urlparse(WEBDAV_BASE_URL)
# Build mapping of sanitized stems to remote URLs
remote_map = {}
from urllib.parse import urlparse, unquote
for url in remote_urls:
if url.startswith('/'):
full_url = f"{parsed_base.scheme}://{parsed_base.netloc}{url}"
else:
full_url = url
# Decode percent-encoding for correct filename comparison
raw_name = unquote(Path(url).name)
stem = Path(raw_name).stem
sanitized_stem = sanitize_filename(stem)
remote_map[sanitized_stem] = full_url
# Build mapping of local stems to file paths
local_map = {Path(p).stem: p for p in file_paths}
# Determine local output directory from first file path
output_dir = Path(file_paths[0]).parent if file_paths else Path('.')
# Download files present remotely but missing locally
for stem, full_url in remote_map.items():
if stem not in local_map:
try:
resp = requests.get(full_url, auth=auth, timeout=30)
if resp.status_code == 200:
new_path = output_dir / f"{stem}.md"
with open(new_path, 'wb') as f:
f.write(resp.content)
print(f"Downloaded remote-only file: {new_path.name}")
file_paths.append(str(new_path))
local_map[stem] = str(new_path)
else:
print(f"Warning: Failed to download remote file {full_url} - Status: {resp.status_code}")
except Exception as e:
print(f"Warning: Could not download {full_url}: {e}")
# Merge existing local and remote content by matching sanitized stems
from difflib import SequenceMatcher
for stem, full_url in remote_map.items():
if stem in local_map:
file_path = local_map[stem]
try:
resp = requests.get(full_url, auth=auth, timeout=30)
if resp.status_code == 200:
# Preserve blank lines when splitting for merge
remote_lines = resp.text.splitlines()
with open(file_path, 'r', encoding='utf-8') as f:
# Preserve blank lines when splitting for merge
local_lines = f.read().splitlines()
matcher = SequenceMatcher(None, local_lines, remote_lines)
merged_lines = []
for tag, i1, i2, j1, j2 in matcher.get_opcodes():
if tag in ('equal', 'delete'):
merged_lines.extend(local_lines[i1:i2])
if tag == 'insert':
merged_lines.extend(remote_lines[j1:j2])
if tag == 'replace':
# Character-level check: if single-line replace
local_block = local_lines[i1:i2]
remote_block = remote_lines[j1:j2]
if len(local_block) == len(remote_block) == 1:
from difflib import SequenceMatcher as CharMatcher
ratio = CharMatcher(None, local_block[0], remote_block[0]).ratio()
# If remote change is minor (ratio>=0.9), keep remote; if major, keep local
if ratio >= 0.9:
merged_lines.append(remote_block[0])
else:
merged_lines.append(local_block[0])
else:
merged_lines.extend(local_block)
merged_lines.extend(remote_block)
# Post-process: collapse near-duplicate lines
from difflib import SequenceMatcher as LineMatcher
filtered = []
for line in merged_lines:
if not filtered:
filtered.append(line)
else:
prev = filtered[-1]
if LineMatcher(None, prev, line).ratio() >= 0.9:
filtered[-1] = line
else:
filtered.append(line)
merged_lines = filtered
# Remove any escape backslashes before square brackets
import re
merged_lines = [re.sub(r'\\([\[\]])', r'\1', l) for l in merged_lines]
with open(file_path, 'w', encoding='utf-8') as f:
f.write('\n'.join(merged_lines))
print(f"Merged file: {Path(file_path).name}")
else:
print(f"Warning: Failed to fetch remote file {full_url} for merging - Status: {resp.status_code}")
except Exception as e:
print(f"Warning: Could not merge {full_url}: {e}")
successful_uploads = []
failed_uploads = []
@@ -569,9 +676,9 @@ def main():
help='Upload the markdown files to WebDAV server'
)
parser.add_argument(
'--no-clear',
'--clear',
action='store_true',
help='Skip clearing WebDAV directory before upload (files will be added/overwritten)'
help='Clear WebDAV directory before upload (default: merge with existing files)'
)
parser.add_argument(
'--direct',
@@ -610,8 +717,8 @@ def main():
# Upload to WebDAV if requested
if args.upload:
# Clear WebDAV directory first (unless --no-clear is specified)
if not args.no_clear:
# Clear WebDAV directory first only if explicitly requested
if args.clear:
if not clear_webdav_directory():
print("Warning: Some files could not be deleted from WebDAV directory.")
print("Continuing with upload (files will be overwritten)...")