winziprint.py: Add feature to abort jobs

To avoid Windows Defender exterminating the .exe file

.gitignore

@@ -1,3 +1,4 @@
 .idea
 __pycache__
 *.exe
+*.spec

CreateExe.bat

@@ -1,6 +1,8 @@
-pyinstaller --noconfirm ^
+pyinstaller ^
+    --noconfirm ^
     --onefile ^
     --console ^
     --icon "NONE" ^
     --name "WinziPrint" ^
+    --clean ^
     src/winziprint.py

README.md

@@ -3,3 +3,43 @@
 
 A standalone [WeasyPrint](https://weasyprint.org/) wrapper for Windows.
 
+
+# Usage
+
+Synopsis:
+```
+winziprint [-h] [-v] [-d DIR] [ -D | [-p] [-2] [-e ENCODING] [ - | INPUT [INPUT...] OUTPUT ] ]
+```
+
+Command line options:
+* `-h`, `--help` - show help message and exit
+* `-V`, `--version` - show version and exit
+* `-D`, `--daemon` - run as a daemon by listening on a TCP socket (port 30983)
+* `-d`, `--directory` - set the working directory
+* `-e`, `--encoding` - encoding of the input files
+* `-2`, `--double-paged` - pad documents to an even number of pages
+* `-p`, `--progress` - show progress updates
+
+Command line arguments:
+* `-` - use stdin for retrieving input and output file names (semi-colon-seperated)
+* `INPUT` - name of a html input file
+* `OUTPUT` - name of a pdf output file
+
+
+## Normal mode
+
+In normal mode all options, input and output files are passed by command line arguments.
+
+
+## Batch mode
+
+To activate batch mode `-` has to be passed as only input or output file.
+In batch mode all options may be passed as command line arguments. These are used as defaults.
+
+
+## Server mode
+
+To activate server mode the `-D` flag has to be set.
+Keep in mind, that not all command line options may be present.
+The protocol is the same as in server mode.
+

src/winziprint.py

@@ -1,73 +1,126 @@
 #!/usr/bin/env python3
 # -*- coding: utf-8 -*-
 
-from typing import Union
+from typing import TextIO, Callable
 import os
 import sys
 import time
 import traceback
 import gc
+import socketserver
+import io
+import signal
+import math
+import threading
 
 import weasyprint
 import pypdf
 
 
-VERSION = __version__ = '0.1.0'
+VERSION = __version__ = '0.3.0'
+SOCKET_ADDRESS = ('127.0.0.1', 30983)
 
 BATCH_SIZE = 10
 
+RUNNING: dict[int, bool] = {}
+
+
+def convert_part(output_file: str, batch: list[str], step_cb: Callable, encoding: str = None) -> list[int]:
+    documents = []
+    for n, file_name in enumerate(batch):
+        html = weasyprint.HTML(filename=file_name, encoding=encoding)
+        doc = html.render()
+        documents.append(doc)
+        del html
+        step_cb()
+    all_pages = [p for doc in documents for p in doc.pages]
+    documents[0].copy(all_pages).write_pdf(output_file)
+    tmp_page_nums = [len(doc.pages) for doc in documents]
+    del documents
+    del all_pages
+    gc.collect()
+    step_cb()
+    return tmp_page_nums
+
 
 def convert(input_files: list[str],
-            output_files: str,
+            output_file: str,
             encoding: str = None,
             padding: bool = False,
-            progress: bool = False) -> list[int]:
+            progress: bool = False,
+            out: TextIO = sys.stdout) -> list[int]:
     # it takes roughly 100ms to generate one document
-    tmp_page_nums = []
     tmp_file_names = []
     page_nums = []
 
-    html_files = [file.lstrip('!') for file in input_files if not file.endswith('.pdf')]
-    steps = len(html_files) + len(html_files) // BATCH_SIZE + 1
+    html_files = [file.lstrip('!#') for file in input_files if not file.endswith('.pdf')]
+    total_steps = len(html_files) + math.ceil(len(html_files) / BATCH_SIZE) + 1
+    steps = [0]
+
+    def next_step() -> None:
+        if not RUNNING[threading.get_ident()]:
+            raise InterruptedError('aborted')
+        steps[0] += 1
+        if progress:
+            print(f'progress: {steps[0]}/{total_steps}', file=out, flush=True)
 
     try:
+        tmp_page_nums = []
         for i in range(0, len(html_files), BATCH_SIZE):
+            tmp_file = f'{output_file}.{i:04}.part'
+            tmp_file_names.append(tmp_file)
             batch = html_files[i:i + BATCH_SIZE]
-            documents = []
-            for n, file_name in enumerate(batch):
-                html = weasyprint.HTML(filename=file_name, encoding=encoding)
-                doc = html.render()
-                documents.append(doc)
-                del html
-                if progress:
-                    print(f'progress: {i + n + i // BATCH_SIZE + 1}/{steps}', flush=True)
-            all_pages = [p for doc in documents for p in doc.pages]
-            tmp_file_name = f'{output_files}.{i:04}.part'
-            documents[0].copy(all_pages).write_pdf(tmp_file_name)
-            tmp_file_names.append(tmp_file_name)
-            tmp_page_nums += [len(doc.pages) for doc in documents]
-            del documents
-            del all_pages
-            gc.collect()
-            if progress and i < BATCH_SIZE:
-                print(f'progress: {i + BATCH_SIZE + i // BATCH_SIZE + 1}/{steps}', flush=True)
+            tmp_page_nums += convert_part(tmp_file, batch, next_step, encoding=encoding)
 
+        letterhead = None
         merger = pypdf.PdfWriter()
         i = 0
         for n, file_name in enumerate(input_files):
             p0 = len(merger.pages)
+            if letterhead and file_name.startswith('#'):
+                if len(merger.pages) <= letterhead[1]:
+                    merger.add_page(letterhead[0])
+                    merger.add_blank_page()
+                else:
+                    merger.insert_page(letterhead[0], index=letterhead[1])
+                    merger.insert_blank_page(index=letterhead[1] + 1)
+                page_nums[letterhead[2]] = 1
+                letterhead = None
             if file_name.endswith('.pdf'):
-                merger.append(file_name.lstrip('!'))
+                if padding and file_name.startswith('#'):
+                    r = pypdf.PdfReader(file_name.lstrip('!#'))
+                    letterhead = (r.pages[0], p0, n)
+                    del r
+                else:
+                    merger.append(file_name.lstrip('!#'))
             else:
                 batch_page_nums = tmp_page_nums[i // BATCH_SIZE * BATCH_SIZE:(i // BATCH_SIZE + 1) * BATCH_SIZE]
                 page_start = sum(batch_page_nums[:i % BATCH_SIZE])
-                merger.append(tmp_file_names[n // BATCH_SIZE], pages=(page_start, page_start + tmp_page_nums[i]))
+                tmp_file_name = tmp_file_names[i // BATCH_SIZE]
+                if padding and file_name.startswith('#'):
+                    r = pypdf.PdfReader(tmp_file_name)
+                    letterhead = (r.pages[page_start], p0, n)
+                    del r
+                else:
+                    merger.append(tmp_file_name, pages=(page_start, page_start + tmp_page_nums[i]))
                 i += 1
             p1 = len(merger.pages)
             page_nums.append(p1 - p0)
-            if padding and file_name[0] != '!' and len(merger.pages) % 2 != 0:
+            if padding and file_name[0] not in ('!', '#') and len(merger.pages) % 2 != 0:
+                if letterhead:
+                    merger.add_page(letterhead[0])
+                    letterhead = None
+                else:
+                    merger.add_blank_page()
+        if letterhead:
+            if len(merger.pages) <= letterhead[1]:
+                merger.add_page(letterhead[0])
                 merger.add_blank_page()
-        merger.write(output_files)
+            else:
+                merger.insert_page(letterhead[0], index=letterhead[1])
+                merger.insert_blank_page(index=letterhead[1] + 1)
+            page_nums[letterhead[2]] = 1
+        merger.write(output_file)
         merger.close()
         del merger
     finally:
@@ -75,52 +128,84 @@ def convert(input_files: list[str],
             if os.path.isfile(pdf):
                 os.remove(pdf)
 
-    if progress:
-        print(f'progress: {steps}/{steps}', flush=True)
+    next_step()
 
     return page_nums
 
 
-def _wrapper_convert(args: list[str], encoding: str = None, padding: bool = False, progress: bool = False) -> None:
+def _wrapper_convert(args: list[str],
+                     encoding: str = None,
+                     padding: bool = False,
+                     progress: bool = False,
+                     out: TextIO = sys.stdout) -> None:
     try:
         if len(args) < 2:
-            print(f'error: Too few arguments', flush=True)
+            print(f'error: Too few arguments', file=out, flush=True)
             return
         inputs = args[:-1]
         output = args[-1]
-        if inputs[0] == '-2':
-            inputs.pop(0)
-            padding = True
+        while len(inputs) > 0:
+            if inputs[0] == '-2':
+                inputs.pop(0)
+                padding = True
+            elif inputs[0].startswith('-e'):
+                encoding = inputs.pop(0)[2:].strip()
+            elif inputs[0].startswith('-p'):
+                inputs.pop(0)
+                progress = True
+            else:
+                break
+        if len(inputs) == 0:
+            print(f'error: Too few arguments', file=out, flush=True)
+            return
         t0 = time.process_time()
-        pages = convert(inputs, output, encoding=encoding, padding=padding, progress=progress)
+        pages = convert(inputs, output, encoding=encoding, padding=padding, progress=progress, out=out)
         total = sum(p + 1 if padding and p % 2 != 0 else p for p in pages)
         t1 = time.process_time()
         print(f'success: '
-              f'{len(args) - 1} documents, '
+              f'{len(inputs)} documents, '
               f'{total} pages ({", ".join(str(p) for p in pages)}), '
               f'{t1 - t0:.1f} sec',
-              flush=True)
+              file=out, flush=True)
     except Exception as e:
         msg = str(e).replace('\n', ' ')
-        print(f'error: {msg}', flush=True)
-        traceback.print_exception(e)
+        print(f'error: {msg}', file=out, flush=True)
+        traceback.print_exception(e, file=sys.stderr)
     finally:
         gc.collect()
 
 
+def tcp_server() -> None:
+    # a tcp server is used due to the lack of unix sockets on Windows
+    with socketserver.ThreadingTCPServer(SOCKET_ADDRESS, ConnectionHandler) as server:
+        def exit_gracefully(_signum: int, _frame) -> None:
+            raise KeyboardInterrupt()
+        signal.signal(signal.SIGINT, exit_gracefully)
+        signal.signal(signal.SIGTERM, exit_gracefully)
+        print('Running as daemon', flush=True)
+        try:
+            server.serve_forever()
+        except KeyboardInterrupt:
+            print('', file=sys.stderr, flush=True)
+    print('Shutting down', flush=True)
+    sys.exit(0)
+
+
 def usage(error: bool = False) -> None:
-    print(f'usage: {sys.argv[0]} [-h] [-v] [-p] [-2] [-d DIR] [-e ENCODING] [ - | INPUT [INPUT...] OUTPUT ]\n\n'
+    print(f'usage: {sys.argv[0]} [-h] [-v] [-d DIR] [ -D | [-p] [-2] [-e ENCODING] [ - | INPUT [INPUT...] OUTPUT ] ]\n'
+          '\n'
           'options:\n'
           ' -h, --help          show this help message and exit\n'
-          ' -v, --version       show version and exit\n'
+          ' -V, --version       show version and exit\n'
+          ' -D, --daemon        run as a daemon and expose a named socket\n'
           ' -d, --directory     set the working directory\n'
           ' -e, --encoding      encoding of the input files\n'
-          ' -2, --double-sided  pad documents to an even number of pages\n'
+          ' -2, --double-paged  pad documents to an even number of pages\n'
           ' -p, --progress      show progress updates\n'
           '\n'
           ' -                   use stdin for retrieving input and output file names (semi-colon-seperated)\n'
-          ' INPUT               name of an html input file\n'
-          ' OUTPUT              name of an pdf output file',
+          ' INPUT               name of a html input file\n'
+          ' OUTPUT              name of a pdf output file',
           file=sys.stderr if error else sys.stdout)
     sys.exit(1 if error else 0)
 
@@ -132,7 +217,7 @@ def version() -> None:
     sys.exit(0)
 
 
-def _get_arg(args: list[str], n1: str, n2: str = None, flag: bool = False) -> Union[None, str, bool]:
+def _get_arg(args: list[str], n1: str, n2: str = None, flag: bool = False) -> None | str | bool:
     v = None
     for n in [n1] + (n2 and [n2] or []):
         if flag:
@@ -150,27 +235,55 @@ def _get_arg(args: list[str], n1: str, n2: str = None, flag: bool = False) -> Un
     return v if not flag else v or False
 
 
+class ConnectionHandler(socketserver.StreamRequestHandler):
+    def handle(self):
+        try:
+            out = io.TextIOWrapper(self.wfile, encoding='utf-8')
+            RUNNING[threading.get_ident()] = True
+            print(f'id: {threading.get_ident()}', file=out, flush=True)
+            for line in io.TextIOWrapper(self.rfile, encoding='utf-8'):
+                if not RUNNING[threading.get_ident()]:
+                    print(f'error: aborted', file=out, flush=True)
+                    break
+                parts = line.strip().split(';')
+                if len(parts) == 2 and parts[0] == 'cancel':
+                    thread_id = int(parts[1].strip())
+                    RUNNING[thread_id] = False
+                    continue
+                _wrapper_convert(parts, out=out)
+        except (ValueError, ConnectionError):
+            pass  # socket closed by client
+        finally:
+            del RUNNING[threading.get_ident()]
+
+
 def main() -> None:
     args = sys.argv[1:]
     if len(args) == 0 or '-h' in args or '--help' in args:
         usage()
-    elif '-v' in args or '--version' in args:
+    elif '-V' in args or '--version' in args:
         version()
 
     working_dir = _get_arg(args, '-d', '--directory')
     if working_dir:
         os.chdir(working_dir)
+
+    if '-D' in args:
+        if len(args) != 1:
+            usage(True)
+        tcp_server()
+
     encoding = _get_arg(args, '-e', '--encoding')
     progress = _get_arg(args, '-p', '--progress', flag=True)
-    double_sided = _get_arg(args, '-2', '--double-sided', flag=True)
+    double_paged = _get_arg(args, '-2', '--double-paged', flag=True)
 
     if args == ['-']:
         for line in sys.stdin:
-            _wrapper_convert(line.strip().split(';'), encoding=encoding, padding=double_sided, progress=progress)
+            _wrapper_convert(line.strip().split(';'), encoding=encoding, padding=double_paged, progress=progress)
     elif len(args) < 2:
         usage(True)
     else:
-        _wrapper_convert(args, encoding=encoding, padding=double_sided, progress=progress)
+        _wrapper_convert(args, encoding=encoding, padding=double_paged, progress=progress)
 
 
 if __name__ == '__main__':