CLI: Use argparse subparsers (#108)

docs/source/changelog.md

@@ -4,6 +4,14 @@
 # Changelog
 
 
+## 1.9.0 (sched 2022-05-16)
+
+- Updated PDFium from `5052` to `XXXX`.
+- Significantly improved maintainability of the command-line interface by using argparse subparsers instead of a custom implementation.
+- Integrated optional support for shell auto-completion using `argcomplete`.
+- Added capabilities to determine bookmark state (open/closed) to the TOC parser. Thanks to PDFium Team for reviewing/merging the CL.
+
+
 ## 1.8.0 (2022-05-09)
 
 - Updated PDFium from `5038` to `5052`.

docs/source/shell_api.rst

@@ -9,41 +9,41 @@ pypdfium2 can also be used from the command-line.
 
 Main help
 *********
-``pypdfium2 help``
+``pypdfium2 --help``
 
-.. program-output:: python3 -m pypdfium2 help
+.. program-output:: pypdfium2 --help
 
 
 Version
 *******
-``pypdfium2 version``
+``pypdfium2 --version``
 
-.. program-output:: python3 -m pypdfium2 version
+.. program-output:: pypdfium2 --version
 
 
 Renderer
 ********
-``pypdfium2 render``
+``pypdfium2 render --help``
 
-.. program-output:: python3 -m pypdfium2 render
+.. program-output:: pypdfium2 render --help
 
 
 Table of Contents Reader
 ************************
-``pypdfium2 toc``
+``pypdfium2 toc --help``
 
-.. program-output:: python3 -m pypdfium2 toc
+.. program-output:: pypdfium2 toc --help
 
 
 Merger
 ******
-``pypdfium2 merge``
+``pypdfium2 merge --help``
 
-.. program-output:: python3 -m pypdfium2 merge
+.. program-output:: pypdfium2 merge --help
 
 
 Page Tiler
 **********
-``pypdfium2 tile``
+``pypdfium2 tile --help``
 
-.. program-output:: python3 -m pypdfium2 tile
+.. program-output:: pypdfium2 tile --help

setup.cfg

@@ -53,10 +53,12 @@ console_scripts =
     pypdfium2 = pypdfium2._cli.main:main
 
 [options.extras_require]
-converters = 
+converters =
     Pillow>=6.0
 inserttext =
     uharfbuzz
+autocomplete =
+    argcomplete
 test =
     pytest
 docs =

src/pypdfium2/_cli/main.py

@@ -1,48 +1,61 @@
+# PYTHON_ARGCOMPLETE_OK
 # SPDX-FileCopyrightText: 2022 geisserml <[email protected]>
 # SPDX-License-Identifier: Apache-2.0 OR BSD-3-Clause
 
 import sys
+import argparse
 from pypdfium2._version import (
     V_PYPDFIUM2,
     V_LIBPDFIUM,
 )
-from pypdfium2._cli._parser import CliParser
 from pypdfium2._cli import (
-    renderer,
+    render,
     toc,
-    merger,
-    tiler,
+    merge,
+    tile,
 )
 
+try:
+    import argcomplete
+except ImportError:
+    have_argcomplete = False
+else:
+    have_argcomplete = True
 
-def main(argv=sys.argv[1:]):
+
+Subcommands = dict(
+    render = render,
+    toc = toc,
+    merge = merge,
+    tile = tile,
+)
+
+
+def parse_args(argv=sys.argv[1:]):
 
-    parser = CliParser(
-        program = "pypdfium2",
-        version = "%s (libpdfium %s)" % (V_PYPDFIUM2, V_LIBPDFIUM),
+    parser = argparse.ArgumentParser(
+        prog = "pypdfium2",
         description = "Command line interface to the pypdfium2 Python library",
-        argv = argv,
-    )
-
-    parser.add_subcommand(
-        "render",
-        method = renderer.main,
-        help = "Rasterise pages of a PDF file",
     )
-    parser.add_subcommand(
-        "toc",
-        method = toc.main,
-        help = "Show the table of contents for a PDF document",
+    parser.add_argument(
+        "--version", "-v",
+        action = "version",
+        version = "pypdfium2 %s (libpdfium %s)" % (V_PYPDFIUM2, V_LIBPDFIUM),
     )
-    parser.add_subcommand(
-        "merge",
-        method = merger.main,
-        help = "Concatenate PDF files",
-    )
-    parser.add_subcommand(
-        "tile",
-        method = tiler.main,
-        help = "Perform page tiling (N-up compositing)",
+
+    subparsers = parser.add_subparsers(
+        dest = "subcommand",
+        required = True,
     )
+    for cmd in Subcommands.values():
+        cmd.attach_parser(subparsers)
+
+    if have_argcomplete:
+        argcomplete.autocomplete(parser)
 
-    parser.run()
+    return parser.parse_args(argv)
+
+
+def main():
+    args = parse_args()
+    Subcommands[args.subcommand].main(args)

src/pypdfium2/_cli/merger.py → src/pypdfium2/_cli/merge.py

@@ -3,7 +3,6 @@
 
 import ctypes
 from pypdfium2 import _namespace as pdfium
-from pypdfium2._cli._parser import ArgParser
 
 
 def _merge_pdfs(input_paths):
@@ -19,13 +18,11 @@ def _merge_pdfs(input_paths):
     return dest_doc
 
 
-def parse_args(argv, prog, desc):
-
-    parser = ArgParser(
-        prog = prog,
-        description = desc,
+def attach_parser(subparsers):
+    parser = subparsers.add_parser(
+        'merge',
+        help = "Concatenate PDF files",
     )
-
     parser.add_argument(
         'inputs',
         nargs = '+',
@@ -36,14 +33,9 @@ def parse_args(argv, prog, desc):
         required = True,
         help = "Target path for the output document",
     )
-
-    return parser.parse_args(argv)
 
 
-def main(argv, prog, desc):
-
-    args = parse_args(argv, prog, desc)
+def main(args):
     merged_doc = _merge_pdfs(args.inputs)
-
     with open(args.output, 'wb') as file_handle:
         pdfium.save_pdf(merged_doc, file_handle)

src/pypdfium2/_cli/renderer.py → src/pypdfium2/_cli/render.py

@@ -4,7 +4,6 @@
 import os
 import ast
 from pypdfium2 import _namespace as pdfium
-from pypdfium2._cli._parser import ArgParser
 from os.path import (
     join,
     abspath,
@@ -69,13 +68,11 @@ def pagetext_type(value):
     return page_indices
 
 
-def parse_args(argv, prog, desc):
-
-    parser = ArgParser(
-        prog = prog,
-        description = desc,
+def attach_parser(subparsers):
+    parser = subparsers.add_parser(
+        "render",
+        help = "Rasterise pages of a PDF file",
     )
-
     parser.add_argument(
         'inputs',
         nargs = '+',
@@ -138,14 +135,10 @@ def parse_args(argv, prog, desc):
         type = int,
         help = "The number of processes to use for rendering (defaults to the number of CPU cores)"
     )
-
-    return parser.parse_args(argv)
 
 
-def main(argv, prog, desc):
-
-    args = parse_args(argv, prog, desc)
-
+def main(args):
+
     for input_path in args.inputs:
 
         prefix = splitext(basename(input_path))[0] + '_'