numbers-parser 4.15.1__tar.gz → 4.16.1__tar.gz

{numbers_parser-4.15.1 → numbers_parser-4.16.1}/PKG-INFO

@@ -1,31 +1,26 @@
-Metadata-Version: 2.3
+Metadata-Version: 2.4
 Name: numbers-parser
-Version: 4.15.1
+Version: 4.16.1
 Summary: Read and write Apple Numbers spreadsheets
-License: MIT
-Author: Jon Connell
-Author-email: python@figsandfudge.com
-Requires-Python: >=3.9,<4.0
-Classifier: License :: OSI Approved :: MIT License
-Classifier: Operating System :: OS Independent
-Classifier: Programming Language :: Python :: 3
-Classifier: Programming Language :: Python :: 3.9
-Classifier: Programming Language :: Python :: 3.10
-Classifier: Programming Language :: Python :: 3.11
-Classifier: Programming Language :: Python :: 3.12
-Classifier: Programming Language :: Python :: 3.13
+Author-email: Jon Connell <python@figsandfudge.com>
+License-Expression: MIT
+Project-URL: repository, https://github.com/masaccio/numbers-parser
+Project-URL: documentation, https://github.com/masaccio/numbers-parser/blob/main/README.md
 Classifier: Topic :: Office/Business :: Financial :: Spreadsheet
-Requires-Dist: compact-json (>=1.1.3,<2.0.0)
-Requires-Dist: enum-tools (>=0.11)
-Requires-Dist: importlib-resources (>=6.1)
-Requires-Dist: protobuf (>=4.0,<6.0)
-Requires-Dist: python-dateutil (>=2.9.0.post0,<3.0.0)
-Requires-Dist: python-snappy (>=0.7,<0.8)
-Requires-Dist: setuptools (>=70.0.0)
-Requires-Dist: sigfig (>=1.3.3,<2.0.0)
-Project-URL: Documentation, https://github.com/masaccio/numbers-parser/blob/main/README.md
-Project-URL: Repository, https://github.com/masaccio/numbers-parser
+Classifier: Programming Language :: Python :: 3
+Classifier: Operating System :: OS Independent
+Requires-Python: <4.0,>=3.9
 Description-Content-Type: text/markdown
+License-File: LICENSE.rst
+Requires-Dist: compact-json<2.0.0,>=1.1.3
+Requires-Dist: protobuf<6.0,>=4.0
+Requires-Dist: python-snappy<1.0,>=0.7
+Requires-Dist: sigfig<2.0.0,>=1.3.3
+Requires-Dist: setuptools>=70.0.0
+Requires-Dist: importlib-resources>=6.1
+Requires-Dist: enum-tools>=0.11
+Requires-Dist: python-dateutil<3.0.0.0,>=2.9.0.post0
+Dynamic: license-file
 
 # numbers-parser
 
@@ -485,8 +480,11 @@ The following limitations are expected to always remain:
 - Password-encrypted documents cannot be opened. You must first re-save without
   a password to read (see [issue 88](https://github.com/masaccio/numbers-parser/issues/88) for details).
   A UnsupportedError exception is raised when such documents are opened.
+- Due to changes in the format of Numbers documents, decoding of category groups
+  (introduced in `numbers-parser` version 4.16) is supported only for documents
+  created by Numbers 12.0 and later. No warnings are issued for earlier
+  Numbers documents.
 
 ## License
 
 All code in this repository is licensed under the [MIT License](https://github.com/masaccio/numbers-parser/blob/master/LICENSE.rst).
-

{numbers_parser-4.15.1 → numbers_parser-4.16.1}/README.md

@@ -456,6 +456,10 @@ The following limitations are expected to always remain:
 - Password-encrypted documents cannot be opened. You must first re-save without
   a password to read (see [issue 88](https://github.com/masaccio/numbers-parser/issues/88) for details).
   A UnsupportedError exception is raised when such documents are opened.
+- Due to changes in the format of Numbers documents, decoding of category groups
+  (introduced in `numbers-parser` version 4.16) is supported only for documents
+  created by Numbers 12.0 and later. No warnings are issued for earlier
+  Numbers documents.
 
 ## License
 

{numbers_parser-4.15.1 → numbers_parser-4.16.1}/pyproject.toml

@@ -1,73 +1,85 @@
-[tool.poetry]
-authors = ["Jon Connell <python@figsandfudge.com>"]
+[project]
+authors = [
+    {name = "Jon Connell", email = "python@figsandfudge.com"},
+]
+license = "MIT"
+requires-python = "<4.0,>=3.9"
+dependencies = [
+    "compact-json<2.0.0,>=1.1.3",
+    "protobuf<6.0,>=4.0",
+    "python-snappy<1.0,>=0.7",
+    "sigfig<2.0.0,>=1.3.3",
+    "setuptools>=70.0.0",
+    "importlib-resources>=6.1",
+    "enum-tools>=0.11",
+    "python-dateutil<3.0.0.0,>=2.9.0.post0",
+]
 classifiers = [
-  "Topic :: Office/Business :: Financial :: Spreadsheet",
-  "Programming Language :: Python :: 3",
-  "Operating System :: OS Independent",
+    "Topic :: Office/Business :: Financial :: Spreadsheet",
+    "Programming Language :: Python :: 3",
+    "Operating System :: OS Independent",
 ]
 description = "Read and write Apple Numbers spreadsheets"
-documentation = "https://github.com/masaccio/numbers-parser/blob/main/README.md"
-license = "MIT"
 name = "numbers-parser"
-packages = [{include = "numbers_parser", from = "src"}]
 readme = "README.md"
+version = "4.16.1"
+
+[project.urls]
 repository = "https://github.com/masaccio/numbers-parser"
-version = "4.15.1"
+documentation = "https://github.com/masaccio/numbers-parser/blob/main/README.md"
 
-[tool.poetry.scripts]
+[project.scripts]
 cat-numbers = "numbers_parser._cat_numbers:main"
 unpack-numbers = "numbers_parser._unpack_numbers:main"
 csv2numbers = "numbers_parser._csv2numbers:main"
 
-[tool.poetry.dependencies]
-compact-json = "^1.1.3"
-protobuf = ">=4.0,<6.0"
-python = ">=3.9,<4.0"
-python-snappy = "^0.7"
-sigfig = "^1.3.3"
-setuptools = ">=70.0.0"
-importlib-resources = ">=6.1"
-enum-tools = ">=0.11"
-python-dateutil = "^2.9.0.post0"
-
-[tool.poetry.group.dev.dependencies]
-gprof2dot = "^2022.7.29"
-line-profiler = "^4.0.3"
-mock = ">=5.1.0"
-pytest = ">=7.2.0"
-pytest-check = ">=1.0"
-pytest-console-scripts = "^1.3.1"
-pytest-cov = ">=4.0,>=5.0"
-pytest-xdist = "^3.3.1"
-ruff = "*"
-tox = "^4.11.4"
-python-magic = ">=0.4"
-tqdm = ">=4.66"
-colorama = "^0.4.6"
-pympler = "^1.1"
-
-[tool.poetry.group.docs]
-optional = true
-
-[tool.poetry.group.docs.dependencies]
-sphinx = ">= 7.3"
-enum-tools = ">=0.11"
-sphinx-toolbox = ">=3.5"
-sphinx-nefertiti = ">=0.7"
-sphinx-markdown-builder = ">=0.6"
-sphinx-copybutton = ">=0.5"
-
-[tool.poetry.group.bootstrap]
-optional = true
-
-[tool.poetry.group.bootstrap.dependencies]
-pyobjc-core = ">=10.2"
-pyobjc-framework-Cocoa = ">=10.2"
-py2app = ">=0.28"
+[dependency-groups]
+dev = [
+    "gprof2dot<2023.0.0,>=2022.7.29",
+    "line-profiler<5.0.0,>=4.0.3",
+    "mock>=5.1.0",
+    "pytest>=7.2.0",
+    "pytest-check>=1.0",
+    "pytest-console-scripts<2.0.0,>=1.3.1",
+    "pytest-cov>=4.0,>=5.0",
+    "pytest-xdist<4.0.0,>=3.3.1",
+    "ruff",
+    "tox<5.0.0,>=4.11.4",
+    "python-magic>=0.4",
+    "tqdm>=4.66",
+    "colorama<1.0.0,>=0.4.6",
+    "pympler<2.0,>=1.1",
+]
+docs = [
+    "sphinx>=7.3",
+    "enum-tools>=0.11",
+    "sphinx-toolbox>=3.5",
+    "sphinx-nefertiti>=0.7",
+    "sphinx-markdown-builder>=0.6",
+    "sphinx-copybutton>=0.5",
+]
+bootstrap = [
+    "pyobjc-core>=10.2",
+    "pyobjc-framework-Cocoa>=10.2",
+    "py2app>=0.28",
+]
 
 [build-system]
-build-backend = "poetry.core.masonry.api"
-requires = ["poetry-core>=1.0.0"]
+requires = ["setuptools >= 61.0"]
+build-backend = "setuptools.build_meta"
+
+# [tool.setuptools]
+# exclude-package-data = { "*" = ["tests*"]}
+# include-package-data = false
+# packages.find.include = ['numbers-parser/src*']
+# packages.find.exclude = ['tests*']
+# package-dir = { "" = "src" }
+
+[tool.setuptools.packages.find]
+where = ["src"]  # list of folders that contain the packages (["."] by default)
+include = ["numbers_parser*"]  # package names should match these glob patterns (["*"] by default)
+exclude = ["*tests*"]  # exclude packages matching these glob patterns (empty by default)
+namespaces = false  # to disable scanning PEP 420 namespaces (true by default)
 
 [tool.coverage.run]
 branch = true
@@ -189,3 +201,4 @@ ban-relative-imports = "all"
 "src/build/protodump.py" = ["PLR2004", "INP001", "PTH", "S110", "N806"]
 "src/debug/**" = ["INP001"]
 "tests/**" = ["PLR2004", "S101", "D103", "ANN201", "ANN001"]
+

numbers_parser-4.16.1/setup.cfg

@@ -0,0 +1,4 @@
+[egg_info]
+tag_build = 
+tag_date = 0
+

numbers_parser-4.16.1/src/build/extract_functions.py

@@ -0,0 +1,104 @@
+import re
+import sys
+from subprocess import PIPE, Popen
+
+# Code pattern in AArch64:
+#
+#         mov     w8, #325
+#         strh    w8, [sp, #8]
+#         str     x23, [sp]
+#         adrp    x2, 2590 ; 0x14e0000
+#         add     x2, x2, #3776 ; Objc cfstring ref: @"GETPIVOTDATA"
+#
+# TSCEFunction_GETPIVOTDATA::evaluateWithContext(...
+
+# Additional required code pattern in AArch64 for Numbers 14.2:
+#
+#         mov     w0, #319
+#         bl      TSCEFormulaCreationMagic::function_3arg(...
+#         ;
+#         ; Approx. 20 lines
+#         ;
+# TSCEFormulaCreationMagic::TEXTBETWEEN(...
+
+if len(sys.argv) != 3:
+    print(f"Usage: {sys.argv[0]} framework-file output.py", file=sys.stderr)
+    sys.exit(1)
+
+framework = sys.argv[1]
+output_map = sys.argv[2]
+
+if framework.endswith(".s"):
+    with open(framework, "rb") as fh:
+        disassembly = fh.readlines()
+else:
+    objdump = Popen(  # noqa: S603
+        [  # noqa: S607
+            "objdump",
+            "--disassemble",
+            "--no-addresses",
+            "--no-print-imm-hex",
+            "--no-show-raw-insn",
+            "--macho",
+            "--objc-meta-data",
+            framework,
+        ],
+        stdout=PIPE,
+    )
+    cxxfilt = Popen(["c++filt"], stdin=objdump.stdout, stdout=PIPE)  # noqa: S603, S607
+    objdump.stdout.close()
+    disassembly = str(cxxfilt.communicate()[0]).split("\\n")
+
+arg = None
+line_count = 0
+tsce_functions = {}
+function_refs = {}
+
+previous_line = ""
+for line in disassembly:
+    line = str(line).replace("\\t", " ")  # noqa: PLW2901
+    if m := re.search(r"mov *w8, #(\d+)", line):
+        arg = m.group(1)
+        line_count = 0
+        continue
+
+    if arg is not None:
+        line_count += 1
+        if line_count > 30:
+            arg = None
+            line_count = 0
+
+    if m := re.search(r'x2, x2.*Objc cfstring ref: @"([A-Z0-9\.]+)"', line):
+        if arg is not None and line_count <= 8:
+            func = m.group(1).replace("_", ".")
+            print(f"Found cstring {func} = {arg}")
+            function_refs[func] = arg
+            arg = None
+            line_count = 0
+    elif m := re.search(r"bl *TSCEFormulaCreationMagic::function_3arg\(", line):
+        if m := re.search(r"mov *w0, #(\d+)", previous_line):
+            arg = m.group(1)
+            line_count = 0
+    elif (m := re.search(r"TSCEFormulaCreationMagic::(\w+)\(", line)) and arg is not None:
+        func = m.group(1).replace("_", ".")
+        print(f"Found TSCEFormulaCreationMagic {func} = {arg}")
+        function_refs[func] = arg
+        arg = None
+        line_count = 0
+    elif m := re.search(r"TSCEFunction_(\w+)::evaluateWithContext", line):
+        func = m.group(1).replace("_", ".")
+        print(f"Found TSCEFunction {func}")
+        tsce_functions[func] = True
+
+    previous_line = line
+
+function_refs = dict(sorted(function_refs.items(), key=lambda x: int(x[1])))
+with open(output_map, "w") as fh:
+    fh.write("FUNCTION_MAP = {\n")
+    if "ABS" not in function_refs:
+        fh.write('    1: "ABS",\n')
+    for func_name, func_id in function_refs.items():
+        if func in tsce_functions:
+            fh.write(f'    {func_id}: "{func_name}",\n')
+
+    fh.write("}\n")

numbers_parser-4.16.1/src/build/extract_mapping.py

@@ -0,0 +1,102 @@
+"""
+lldb script to dump TSPersistence registry mapping from iWork apps.
+
+Licensed under the MIT license.
+
+Copyright 2020-2022 Peter Sobot
+Copyright 2022 Jon Connell
+Copyright 2022 SheetJS LLC
+"""
+
+import json
+import os
+import sys
+
+import lldb
+
+from debug.lldbutil import print_stacktrace
+
+if len(sys.argv) != 3:
+    msg = f"Usage: {sys.argv[0]} exe-file output.json"
+    raise (ValueError(msg))
+
+exe = sys.argv[1]
+output = sys.argv[2]
+
+debugger = lldb.SBDebugger.Create()
+debugger.SetAsync(False)
+target = debugger.CreateTargetWithFileAndArch(exe, None)
+
+# # Note: original script also created breakpoints on _handleAEOpenEvent
+# # but that is too early in Numbers 12.1
+# target.BreakpointCreateByName("-[NSApplication _sendFinishLaunchingNotification]")
+# target.BreakpointCreateByName("-[NSApplication _crashOnException:]")
+
+# # Note: original script skipped [CKContainer containerWithIdentifier:]
+# target.BreakpointCreateByRegex("CloudKit")
+
+target.BreakpointCreateByName("_sendFinishLaunchingNotification")
+target.BreakpointCreateByName("_handleAEOpenEvent:")
+# To get around the fact that we don't have iCloud entitlements when running re-signed code,
+# let's break in the CloudKit code and early exit the function before it can raise an exception:
+target.BreakpointCreateByName("[CKContainer containerWithIdentifier:]")
+# In later Keynote versions, 'containerWithIdentifier' isn't called directly, but we can break on similar methods:
+# Note: this __lldb_unnamed_symbol hack was determined by painstaking experimentation. It will break again for sure.
+target.BreakpointCreateByRegex("___lldb_unnamed_symbol[0-9]+", "CloudKit")
+
+
+process = target.LaunchSimple(None, None, os.getcwd())
+if not process:
+    raise ValueError("Failed to launch process: " + exe)
+
+if process.GetState() == lldb.eStateExited:
+    msg = f"LLDB was unable to stop process! {process}"
+    raise ValueError(msg)
+
+try:
+    while process.GetState() == lldb.eStateStopped:
+        thread = process.GetThreadAtIndex(0)
+        frame = thread.GetSelectedFrame()
+        if frame.name == "-[NSApplication _crashOnException:]":
+            msg = f"Process crashed at {frame.name}"
+            raise ValueError(msg)
+
+        stop_reason = thread.GetStopReason()
+
+        if stop_reason == lldb.eStopReasonException:
+            print_stacktrace(thread)
+            function = frame.GetFunction()
+            function_or_symbol = function if function else frame.GetSymbol()
+            msg = f"Exception at {frame.name}"
+            raise ValueError(msg)
+        if stop_reason != lldb.eStopReasonBreakpoint:
+            process.Continue()
+            continue
+        if frame.name[-8:] == "CloudKit":
+            thread.ReturnFromFrame(
+                thread.GetSelectedFrame(),
+                lldb.SBValue().CreateValueFromExpression("0", ""),
+            )
+            process.Continue()
+        elif frame.name == "-[NSApplication _sendFinishLaunchingNotification]":
+            registry = frame.EvaluateExpression("[TSPRegistry sharedRegistry]")
+            error = registry.GetError()
+            if error.fail or registry.description is None:
+                continue
+                # raise (ValueError("Failed to extract registry"))
+            split = [
+                x.strip().split(" -> ")
+                for x in registry.description.split("{")[1].split("}")[0].split("\n")
+                if x.strip()
+            ]
+            json_str = json.dumps(
+                dict(sorted([(int(a), b.split(" ")[-1]) for a, b in split if "null" not in b])),
+                indent=2,
+            )
+            with open(output, "w") as fh:
+                fh.write(json_str)
+            break
+        else:
+            process.Continue()
+finally:
+    process.Kill()

numbers_parser-4.16.1/src/build/generate_fontmap.py

@@ -0,0 +1,22 @@
+import sys
+
+import Cocoa
+
+if len(sys.argv) != 2:
+    msg = f"Usage: {sys.argv[0]} fontmap.py"
+    raise (ValueError(msg))
+
+mapping_py = sys.argv[1]
+
+with open(mapping_py, "w") as fh:
+    manager = Cocoa.NSFontManager.sharedFontManager()
+    font_families = list(manager.availableFontFamilies())
+
+    print("FONT_NAME_TO_FAMILY = {", file=fh)
+    for family in sorted(font_families):
+        fonts = manager.availableMembersOfFontFamily_(family)
+        for font in fonts:
+            print(f'    "{font[0]}": "{family}",', file=fh)
+    print('    "Calibri": "Calibri",', file=fh)
+    print('    "Cambria": "Cambria",', file=fh)
+    print("}", file=fh)

numbers_parser-4.16.1/src/build/generate_mapping.py

@@ -0,0 +1,70 @@
+import json
+import os
+import sys
+
+if len(sys.argv) != 3:
+    msg = f"Usage: {sys.argv[0]} mapping.json mapping.py"
+    raise (ValueError(msg))
+
+mapping_json = sys.argv[1]
+mapping_py = sys.argv[2]
+
+modules = []
+module_import_output = ""
+module_files_output = ""
+mapping_output = ""
+for filename in os.listdir("src/numbers_parser/generated"):
+    if "pb2" in filename:
+        module = filename.replace("_pb2.py", "")
+        modules.append(module)
+        module_import_output += f"from numbers_parser.generated import {module}_pb2 as {module}\n"
+        module_files_output += f"    {module},\n"
+
+with open(mapping_json) as fh:
+    mappings = json.load(fh)
+
+for index, symbol in sorted(mappings.items(), key=lambda x: int(x[0])):
+    mapping_output += f'    "{index}": "{symbol}",\n'
+
+OUTPUT_CODE = f"""{module_import_output.strip()}
+
+PROTO_FILES = [
+    {module_files_output.strip()}
+]
+
+TSPRegistryMapping = {{
+    {mapping_output.strip()}
+}}
+
+
+def compute_maps():
+    name_class_map = {{}}
+
+    def add_nested_types(message_type):
+        for name in dict(message_type.DESCRIPTOR.nested_types_by_name):
+            child_type = getattr(message_type, name)
+            name_class_map[child_type.DESCRIPTOR.full_name] = child_type
+            add_nested_types(child_type)
+
+    for file in PROTO_FILES:
+        for message_name in dict(file.DESCRIPTOR.message_types_by_name):
+            message_type = getattr(file, message_name)
+            name_class_map[message_type.DESCRIPTOR.full_name] = message_type
+            add_nested_types(message_type)
+
+    id_name_map = {{}}
+    name_id_map = {{}}
+    for k, v in list(TSPRegistryMapping.items()):
+        if v in name_class_map:  # pragma: no branch
+            id_name_map[int(k)] = name_class_map[v]
+            if v not in name_id_map:
+                name_id_map[v] = int(k)
+
+    return name_class_map, id_name_map, name_id_map
+
+
+NAME_CLASS_MAP, ID_NAME_MAP, NAME_ID_MAP = compute_maps()
+"""
+
+with open(mapping_py, "w") as fh:
+    fh.write(OUTPUT_CODE)