Ready for 1.0!

Neaten up the command line interface, rip out some bad ideas, and push
with some documentation.

README.md

@@ -1 +1,75 @@
-`macresources` readme
+# macresources
+
+A Python library and command line tools to work with Classic MacOS [resource
+forks](https://en.wikipedia.org/wiki/Resource_fork) on a modern machine.
+
+
+## Data Format
+
+First, `macresources` and its sister package
+[`machfs`](https://pypi.org/project/machfs/) have a preferred representation for
+Macintosh files, where Macintosh-specific information is stored in separate text
+files.
+
+1. The data fork is stored inside a file with the original name. This must be
+present for the following two files to be recognised.
+
+2. The resource fork is stored in a 'Rez-style' textfile with `.rdump` appended
+to the original name. The format is slightly different from a vanilla 'DeRez'
+dump: non-ASCII characters are escaped, giving an ASCII-clean output:
+
+        data '\0x96tbl' (0) {
+            $"0000 0001 0000 0000 0000 0010 0669 4D61"            /* .............iMa */
+            ...
+        };
+
+3. The four-character type and creator codes are concatenated (like a `PkgInfo`
+file inside an app bundle) in a file with `.idump` appended to the original
+name. If the type is `TEXT` or `ttro`, then the data fork is converted to UTF-8
+with Unix (LF) line endings.
+
+Several other formats exist to store this Macintosh specific data in flat files,
+the best known being
+[AppleSingle/AppleDouble](https://en.wikipedia.org/wiki/AppleSingle_and_AppleDouble_formats),
+[MacBinary](https://en.wikipedia.org/wiki/MacBinary) and
+[BinHex 4](https://en.wikipedia.org/wiki/BinHex). The data format described here
+instead adapts text-friendly formats (`Rez` and `PkgInfo`). The result is
+especially useful for placing legacy Macintosh source code under modern version
+control.
+
+The role of `macresources` is to produce and parse Rez-style `.rdump` files, and
+to produce and parse raw resource forks for `machfs` disk images.
+
+
+## Command Line Interface
+
+`rfx` is a shell command wrapper for accessing resources inside a `.rdump` file.
+Command line arguments are passed through to the command, but resources
+specified as `filename.rdump//type/id` are converted to tempfiles before the
+command is run, and back to resources after the command returns. This approach
+even enables `cp`, `mv` and `rm` to work on individual resources.
+
+`rezhex` and `hexrez` convert between
+[BinHex](https://en.wikipedia.org/wiki/BinHex) (`.hqx`) format and
+`macresources`/`macbinary` format.
+
+`SimpleRez` and `SimpleDeRez` are very simple reimplementations of the
+deprecated `Rez` and `DeRez` utilities. They convert between raw resource forks
+and Rez-style `.rdump` files. To access a raw resource fork under Mac OS X, you
+can append `/..namedfork/rsrc` to a filename.
+
+All utilities have online help.
+
+
+## API
+
+The Python API is pretty spartan. It exists mainly to support `machfs` and the command line interface.
+
+    from macresources import *
+
+    make_rez_code(from_iter, ascii_clean=False)     # Takes an iterator of Resource objects, returns Rez code
+    parse_rez_code(from_code)                       # Takes Rez code, returns an iterator of Resource objects
+    make_file(from_iter)                            # Takes an iterator of Resource objects, returns a raw resource fork
+    parse_file(from_file)                           # Takes a raw resource fork, returns an iterator of Resource objects
+
+The `Resource` class inherits from bytearray.

bin/Rget

@@ -1,76 +0,0 @@
-#!/usr/bin/env python3
-
-import argparse
-import macresources
-import sys
-import tempfile
-from os import path
-
-def fourcc(s):
-    b = s.encode('mac_roman').ljust(4, b' ')
-    if len(b) != 4:
-        raise ValueError('wrong length')
-    return b
-
-def seemshex(s):
-    s = s.lower()
-    if not s: return False
-    return all(c in '0123456789abcdef' for c in s)
-
-def seemsdec(s):
-    if not s: return False
-    return all(c in '0123456789' for c in s)
-
-def resid(s):
-    s = s.lower()
-    if s.startswith('0x') and seemshex(s[2:]):
-        thenum = int(s[2:], 16)
-        if thenum > 0x7fff: thenum -= 0x8000
-    elif s.startswith('$') and seemshex(s[1:]):
-        thenum = int(s[1:], 16)
-        if thenum > 0x7fff: thenum -= 0x8000
-    else:
-        thenum = int(s)
-    if not (-0x8000 <= thenum <= 0x7fff): raise ValueError
-    return thenum
-
-parser = argparse.ArgumentParser(description='''
-    Copy a single MacOS resource from a source file to the standard
-    output. If the source filename ends with `.r' or `.rdump', then it
-    is parsed using the `SimpleRez' subset of the Rez language. TODO:
-    find a way to specify null characters so that desk accessories can
-    be looked up by name.
-''')
-
-parser.add_argument('srcfile', help='resource file or Rez file')
-parser.add_argument('type', type=fourcc, help='four-byte type of resource (converted to Mac Roman pre-lookup)')
-parser.add_argument('id', type=resid, help='ID number of resource (-32768 to 32767)')
-parser.add_argument('-f', dest='tofile', action='store_true', help='copy to a tempfile instead')
-
-args = parser.parse_args()
-
-with open(args.srcfile, 'rb') as f:
-    raw = f.read()
-
-if args.srcfile.endswith('.r') or args.srcfile.endswith('.rdump'):
-    resources = macresources.parse_rez_code(raw)
-else:
-    resources = macresources.parse_file(raw)
-
-for r in resources:
-    if r.type == args.type and r.id == args.id:
-        myres = r
-        break
-else:
-    raise ValueError(args.type, args.id)
-
-if args.tofile:
-    tmpname = '-%s-%s-%d' % (path.basename(args.srcfile), myres.type.decode('mac_roman'), myres.id)
-    with tempfile.NamedTemporaryFile(suffix=tmpname, delete=False, mode='wb') as f:
-        f.write(myres.data)
-        print(f.name)
-else:
-    try:
-        sys.stdout.buffer.write(myres.data)
-    except BrokenPipeError:
-        pass

bin/Rhqx

@@ -1,80 +0,0 @@
-import os
-from os import path
-import argparse
-import macresources
-from macresources import binhex
-
-parser = argparse.ArgumentParser(description='''
-    Supply base name to convert datafork+rdump+idump to HQX.
-    Supply base.hqx name to convert HQX to datafork+rdump+idump.
-''')
-
-parser.add_argument('srcfile', nargs='*', help='base or base.hqx')
-
-args = parser.parse_args()
-
-
-for srcfile in args.srcfile:
-    # Case 1: from BinHex
-    if path.splitext(path.basename(srcfile))[1].lower() == '.hqx':
-        hb = binhex.HexBin(srcfile)
-
-        base = path.splitext(srcfile)[0]
-
-        if hb.FInfo.Type == hb.FInfo.Creator == b'????':
-            try:
-                os.remove(base + '.idump')
-            except FileNotFoundError:
-                pass
-        else:
-            with open(base + '.idump', 'wb') as f:
-                f.write(hb.FInfo.Type + hb.FInfo.Creator)
-
-        data = hb.read()
-        if hb.FInfo.Type in [b'TEXT', b'ttro']:
-            data = data.replace(b'\r', b'\n').decode('mac_roman').encode('utf-8')
-        with open(base, 'wb') as f:
-            f.write(data)
-
-        rsrc = hb.read_rsrc()
-        if rsrc:
-            with open(base + '.rdump', 'wb') as f:
-                f.write(macresources.make_rez_code(macresources.parse_file(rsrc), ascii_clean=True))
-        else:
-            try:
-                os.remove(base + '.rdump')
-            except FileNotFoundError:
-                pass
-
-    # Case 2: to BinHex
-    else:
-        finfo = binhex.FInfo()
-        finfo.Flags = 0
-
-        try:
-            info = open(srcfile + '.idump', 'rb').read(8)
-            assert len(info) == 8
-            finfo.Type = info[:4]
-            finfo.Creator = info[4:]
-        except:
-            pass
-
-        try:
-            data = open(srcfile, 'rb').read()
-            if finfo.Type in [b'TEXT', b'ttro']:
-                data = data.replace(b'\n', b'\r').decode('utf-8').encode('mac_roman')
-        except:
-            data = b''
-
-        try:
-            rsrc = open(srcfile + '.rdump', 'rb').read()
-            rsrc = macresources.make_file(macresources.parse_rez_code(rsrc))
-        except:
-            rsrc = b''
-
-        bh = binhex.BinHex((path.basename(srcfile), finfo, len(data), len(rsrc)), srcfile + '.hqx')
-
-        bh.write(data)
-        bh.write_rsrc(rsrc)
-
-        bh.close()

bin/hexrez

@@ -0,0 +1,70 @@
+#!/usr/bin/env python3
+
+import os
+from os import path
+import argparse
+import macresources
+from macresources import binhex
+
+
+def do_file(the_path):
+    base_path = path.splitext(the_path)[0] # known to have hqx extension
+    hb = binhex.HexBin(the_path)
+
+    if hb.FInfo.Type == hb.FInfo.Creator == b'????':
+        try:
+            os.remove(base_path + '.idump')
+        except FileNotFoundError:
+            pass
+    else:
+        with open(base_path + '.idump', 'wb') as f:
+            f.write(hb.FInfo.Type + hb.FInfo.Creator)
+
+    data = hb.read()
+    if hb.FInfo.Type in [b'TEXT', b'ttro']:
+        data = data.replace(b'\r', b'\n').decode('mac_roman').encode('utf-8')
+    with open(base_path, 'wb') as f:
+        f.write(data)
+
+    rsrc = hb.read_rsrc()
+    if rsrc:
+        with open(base_path + '.rdump', 'wb') as f:
+            f.write(macresources.make_rez_code(macresources.parse_file(rsrc), ascii_clean=True))
+    else:
+        try:
+            os.remove(base_path + '.rdump')
+        except FileNotFoundError:
+            pass
+
+
+def is_hqx_name(the_path):
+    name = path.basename(the_path)
+    base, ext = path.splitext(name)
+    if ext.lower() == '.hqx':
+        return True
+    else:
+        return False
+
+
+parser = argparse.ArgumentParser(description='''
+    UnBinHex (BASE.hqx) into (BASE + BASE.rdump + BASE.idump)
+''')
+
+parser.add_argument('hqx', metavar='BASE.hqx', nargs='+', help='file or directory')
+
+args = parser.parse_args()
+
+for hqx in args.hqx:
+    if path.isdir(hqx):
+        for hqx, dirlist, filelist in os.walk(hqx):
+            dirlist[:] = [d for d in dirlist if not d.startswith('.')]; dirlist.sort()
+            filelist[:] = [f for f in filelist if not f.startswith('.')]; filelist.sort()
+
+            for f in filelist:
+                if is_hqx_name(f):
+                    do_file(path.join(hqx, f))
+    else:
+        if not is_hqx_name(hqx):
+            exit('Not a BinHex file')
+
+        do_file(hqx)

bin/rezhex

@@ -0,0 +1,72 @@
+#!/usr/bin/env python3
+
+import os
+from os import path
+import argparse
+import macresources
+from macresources import binhex
+
+
+def do_file(the_path):
+    finfo = binhex.FInfo()
+    finfo.Flags = 0
+
+    try:
+        info = open(the_path + '.idump', 'rb').read(8)
+        assert len(info) == 8
+        finfo.Type = info[:4]
+        finfo.Creator = info[4:]
+    except:
+        pass
+
+    try:
+        data = open(the_path, 'rb').read()
+        if finfo.Type in [b'TEXT', b'ttro']:
+            data = data.replace(b'\n', b'\r').decode('utf-8').encode('mac_roman')
+    except:
+        data = b''
+
+    try:
+        rsrc = open(the_path + '.rdump', 'rb').read()
+        rsrc = macresources.make_file(macresources.parse_rez_code(rsrc))
+    except:
+        rsrc = b''
+
+    bh = binhex.BinHex((path.basename(the_path), finfo, len(data), len(rsrc)), the_path + '.hqx')
+
+    bh.write(data)
+    bh.write_rsrc(rsrc)
+
+    bh.close()
+
+
+def is_valid_base(the_path):
+    name = path.basename(the_path)
+    base, ext = path.splitext(name)
+    if ext.lower() in ('.hqx', '.idump', '.rdump'): return False
+    return True
+
+
+parser = argparse.ArgumentParser(description='''
+    BinHex (BASE + BASE.rdump + BASE.idump) into (BASE.hqx)
+''')
+
+parser.add_argument('base', metavar='BASE', nargs='+', help='file or directory')
+
+args = parser.parse_args()
+
+for base in args.base:
+    if path.isdir(base):
+        for base, dirlist, filelist in os.walk(base):
+            dirlist[:] = [d for d in dirlist if not d.startswith('.')]; dirlist.sort()
+            filelist[:] = [f for f in filelist if not f.startswith('.')]; filelist.sort()
+
+            for f in filelist:
+                if is_valid_base(f):
+                    do_file(path.join(base, f))
+
+    else:
+        if not is_valid_base(hqx):
+            exit('Base names cannot have a .hqx/.idump/.rdump extension')
+
+        do_file(base)

bin/rfx

@@ -0,0 +1,146 @@
+#!/usr/bin/env python3
+
+import macresources
+import sys
+import tempfile
+from os import path
+import re
+import subprocess
+import textwrap
+
+
+if len(sys.argv) < 2 or sys.argv[1].startswith('-'):
+    sys.exit(textwrap.dedent('''
+        usage: rfx command [arg[//type/id] ...]
+
+        Shell command wrapper for accessing resources inside a Rez textfile
+
+        Resources specified as filename.rdump//type/id are converted to tempfiles before
+        the command is run, and back to resources after the command returns.
+
+        examples:
+            rfx mv Doc.rdump//STR/0 Doc.rdump//STR/1
+            rfx cp App.rdump//PICT/2000 2000.pict
+            rfx rm System.rdump//vers/2
+    ''').strip())
+
+
+bytearray_cache = {}
+original_cache = {}
+
+def get_cached_file(the_path):
+    # Different paths to the same file are unlikely, but just in case:
+    the_path = path.abspath(the_path)
+
+    try:
+        return bytearray_cache[the_path]
+    except KeyError:
+        try:
+            with open(the_path, 'rb') as f:
+                d = f.read()
+        except FileNotFoundError:
+            d = bytes()
+
+        original_cache[the_path] = d
+        bytearray_cache[the_path] = bytearray(d)
+        return bytearray_cache[the_path]
+
+def flush_cache():
+    for the_path, the_data in bytearray_cache.items():
+        if original_cache[the_path] != the_data:
+            with open(the_path, 'wb') as f:
+                f.write(the_data)
+
+
+def rez_resource_range(the_data, the_type, the_id):
+    if not the_data: return (0, 0)
+
+    # Hack... do a text search instead of Rezzing the whole file!
+    search = macresources.make_rez_code([macresources.Resource(the_type, the_id)], ascii_clean=True)
+    search = search.rpartition(b')')[0]
+
+    start = 0
+    while True:
+        start = the_data.find(search, start)
+        if start == -1: return (0, 0)
+        if (the_data[start-1:start] in b'\n') and (the_data[start+len(search):start+len(search)+1] in (b',', b')')):
+            break
+        start += len(search)
+
+    stop = the_data.index(b'\n};\n\n', start) + 5
+
+    return (start, stop)
+
+
+def rez_shrink_range(the_data, start, stop):
+    start = the_data.index(b'\n', start) + 1
+    while the_data[stop:stop+1] != b'}': stop -= 1
+
+    return (start, stop)
+
+
+def rez_get_resource(the_path, the_type, the_id):
+    the_file = get_cached_file(the_path)
+
+    start, stop = rez_resource_range(the_file, the_type, the_id)
+    if start == stop == 0: return None
+    return next(macresources.parse_rez_code(the_file[start:stop])).data
+
+
+def rez_set_resource(the_path, the_type, the_id, the_data):
+    the_file = get_cached_file(the_path)
+
+    newdata = macresources.make_rez_code([macresources.Resource(the_type, the_id, data=the_data)], ascii_clean=True)
+
+    start, stop = rez_resource_range(the_file, the_type, the_id)
+    if start == stop == 0:
+        the_file.extend(newdata)
+    else:
+        start, stop = rez_shrink_range(the_file, start, stop)
+        istart, istop = rez_shrink_range(newdata, 0, len(newdata))
+
+        the_file[start:stop] = newdata[istart:istop]
+
+
+def rez_delete_resource(the_path, the_type, the_id):
+    the_file = get_cached_file(the_path)
+
+    start, stop = rez_resource_range(the_file, the_type, the_id)
+    del the_file[start:stop]
+
+
+with tempfile.TemporaryDirectory() as backup_tmp_dir:
+    new_argv = [sys.argv[1]]
+    to_retrieve = []
+
+    for i, arg in enumerate(sys.argv[2:], 1):
+        m = re.match(r'(.*[^/])//([^/]{1,4})/(-?\d+)$'.replace('/', re.escape(path.sep)), arg)
+
+        if m:
+            res_spec = (m.group(1), m.group(2).encode('mac_roman').ljust(4)[:4], int(m.group(3)))
+            tmp_file = path.join(backup_tmp_dir, str(i))
+
+            to_retrieve.append((tmp_file, res_spec))
+
+            res_data = rez_get_resource(*res_spec)
+            if res_data is not None:
+                with open(tmp_file, 'wb') as f:
+                    f.write(res_data)
+
+            new_argv.append(tmp_file)
+
+        else:
+            new_argv.append(arg)
+
+    result = subprocess.run(new_argv)
+
+    for tmp_file, res_spec in to_retrieve:
+        try:
+            with open(tmp_file, 'rb') as f:
+                rez_set_resource(*res_spec, f.read())
+        except FileNotFoundError:
+            rez_delete_resource(*res_spec)
+
+flush_cache()
+
+sys.exit(result.returncode)

bin/sortrez

@@ -0,0 +1,68 @@
+#!/usr/bin/env python3
+
+import argparse
+import macresources
+
+
+MACROMAN_SORT = [
+    0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07,
+    0x08, 0x09, 0x0a, 0x0b, 0x0c, 0x0d, 0x0e, 0x0f,
+    0x10, 0x11, 0x12, 0x13, 0x14, 0x15, 0x16, 0x17,
+    0x18, 0x19, 0x1a, 0x1b, 0x1c, 0x1d, 0x1e, 0x1f,
+
+    0x20, 0x22, 0x23, 0x28, 0x29, 0x2a, 0x2b, 0x2c,
+    0x2f, 0x30, 0x31, 0x32, 0x33, 0x34, 0x35, 0x36,
+    0x37, 0x38, 0x39, 0x3a, 0x3b, 0x3c, 0x3d, 0x3e,
+    0x3f, 0x40, 0x41, 0x42, 0x43, 0x44, 0x45, 0x46,
+
+    0x47, 0x48, 0x58, 0x5a, 0x5e, 0x60, 0x67, 0x69,
+    0x6b, 0x6d, 0x73, 0x75, 0x77, 0x79, 0x7b, 0x7f,
+    0x8d, 0x8f, 0x91, 0x93, 0x96, 0x98, 0x9f, 0xa1,
+    0xa3, 0xa5, 0xa8, 0xaa, 0xab, 0xac, 0xad, 0xae,
+
+    0x54, 0x48, 0x58, 0x5a, 0x5e, 0x60, 0x67, 0x69,
+    0x6b, 0x6d, 0x73, 0x75, 0x77, 0x79, 0x7b, 0x7f,
+    0x8d, 0x8f, 0x91, 0x93, 0x96, 0x98, 0x9f, 0xa1,
+    0xa3, 0xa5, 0xa8, 0xaf, 0xb0, 0xb1, 0xb2, 0xb3,
+
+    0x4c, 0x50, 0x5c, 0x62, 0x7d, 0x81, 0x9a, 0x55,
+    0x4a, 0x56, 0x4c, 0x4e, 0x50, 0x5c, 0x62, 0x64,
+    0x65, 0x66, 0x6f, 0x70, 0x71, 0x72, 0x7d, 0x89,
+    0x8a, 0x8b, 0x81, 0x83, 0x9c, 0x9d, 0x9e, 0x9a,
+
+    0xb4, 0xb5, 0xb6, 0xb7, 0xb8, 0xb9, 0xba, 0x95,
+    0xbb, 0xbc, 0xbd, 0xbe, 0xbf, 0xc0, 0x52, 0x85,
+    0xc1, 0xc2, 0xc3, 0xc4, 0xc5, 0xc6, 0xc7, 0xc8,
+    0xc9, 0xca, 0xcb, 0x57, 0x8c, 0xcc, 0x52, 0x85,
+
+    0xcd, 0xce, 0xcf, 0xd0, 0xd1, 0xd2, 0xd3, 0x26,
+    0x27, 0xd4, 0x20, 0x4a, 0x4e, 0x83, 0x87, 0x87,
+    0xd5, 0xd6, 0x24, 0x25, 0x2d, 0x2e, 0xd7, 0xd8,
+    0xa7, 0xd9, 0xda, 0xdb, 0xdc, 0xdd, 0xde, 0xdf,
+
+    0xe0, 0xe1, 0xe2, 0xe3, 0xe4, 0xe5, 0xe6, 0xe7,
+    0xe8, 0xe9, 0xea, 0xeb, 0xec, 0xed, 0xee, 0xef,
+    0xf0, 0xf1, 0xf2, 0xf3, 0xf4, 0xf5, 0xf6, 0xf7,
+    0xf8, 0xf9, 0xfa, 0xfb, 0xfc, 0xfd, 0xfe, 0xff,
+]
+
+
+def sortkey(resource):
+    return (*(MACROMAN_SORT[char] for char in resource.type), resource.id)
+
+
+parser = argparse.ArgumentParser(description='''
+    Sort the resources in a Rez file (for diffing).
+''')
+
+parser.add_argument('src', nargs='*', help='Rez files')
+args = parser.parse_args()
+
+for srcfile in args.src:
+    with open(srcfile, 'r+b') as f:
+        raw = f.read()
+        resources = list(macresources.parse_rez_code(raw))
+        resources.sort(key=sortkey)
+        f.seek(0)
+        f.truncate(0)
+        f.write(macresources.make_rez_code(resources, ascii_clean=True))

setup.py

@@ -2,7 +2,7 @@ from setuptools import setup
 
 setup(
     name='macresources',
-    version='0.1dev',
+    version='1.0',
     author='Elliot Nunn',
     author_email='elliotnunn@me.com',
     description='Library for working with legacy Macintosh resource forks',
@@ -18,5 +18,5 @@ setup(
         'Development Status :: 3 - Alpha',
     ],
     packages=['macresources'],
-    scripts=['bin/SimpleRez', 'bin/SimpleDeRez', 'bin/Rget', 'bin/Rhqx'],
+    scripts=['bin/SimpleRez', 'bin/SimpleDeRez', 'bin/hexrez', 'bin/rezhex', 'bin/sortrez', 'bin/rfx'],
 )