Notes on Radmind’s checksum

Leave a reply

It would be nice to do a [pure-Python][python] implementation of [Radmind][radmind]’s fsdiff output for [watchedinstall][watchedinstall], which consists of several white-space separated fields describing the filename’s attributes and an optional checksum for the file.

These are notes on how Radmind generates checksums for files on [Mac OS X][macosx].

The [fsdiff format is documented][manfsdiff], however for files with Mac Finder info or a resource fork the checksum is for an [AppleSingle][applesingle]-encoded representation of the file, which means a Python implementation needs to produce an equivalent AppleSingle-encoded byte stream for the file. Bummer.

Python 2.6 on Mac OS X includes a [(deprecated) applesingle module][applesinglemod] that can read the format but cannot write it (and the module has been removed for Python 3). Therefore a pure Python implementation of Radmind’s checksum has to implement a compatible AppleSingle encoding routine too.

Radmind’s fsdiff command is written in C, which I can just about get the gist of, but I am missing something because my attempts at emulating Radmind’s checksums are wrong.

The meat of Radmind’s checksum is the [`do_acksum()` function in `cksum.c`][cksum]. The algorithm appears to be as follows:

1. Initialize a digest using the default cipher ([MD5][md5] I think).
2. Add the AppleSingle header, consisting of a magic number and version number and some padding.
3. Add the AppleSingle entry table, which has 3 entries for the Finder info, the resource fork info and the data fork info (in that order). Each entry is 12 bytes – an unsigned long for the entry type, an unsigned long for an offset into the file where the data will start and an unsigned long for the data length.
4. Add the Finder info data.
5. Add the resource for data.
6. Add the data fork data.
7. Return a base64 encoded version of the final digest.

Because the entry table in the AppleSingle header specifies data offsets and lengths you need to know the size of the Finder info data (always 32 bytes) and the size of the resource fork and the size of the data fork before you pass that data to the digest function.

So a working Python implementation needs to know the size of the resource fork and data fork before feeding that same data to the digest. It seems to me that this requirement might imply huge memory allocations while slurping file data – my wrong attempt tried counting bytes and later feeding the same data to the digest in manageable chunks.

Anyway…

Advice much appreciated. The workaround is to leave it to fsdiff to generate the checksum and parse the value from the output.

David

P.S. I still intend running [A/UX 3.0.1][aux] on my Centris 660av one day.

Update: using my eyes and brains and the `fsdiff -V` command I was able to read the fsdiff man page and deduce the preferred checksum cipher is actually sha1. My code is still wrong.

[radmind]: http://rsug.itd.umich.edu/software/radmind/
[python]: http://www.python.org/
[macosx]: http://www.apple.com/macosx/
[manfsdiff]: http://linux.die.net/man/1/fsdiff
[applesingle]: http://users.phg-online.de/tk/netatalk/doc/Apple/v2/AppleSingle_AppleDouble.pdf
[applesinglemod]: http://www.python.org/doc/2.6.2/library/undoc.html#module-applesingle
[md5]: http://www.openssl.org/docs/crypto/md5.html
[aux]: http://www.aux-penelope.com/
[cksum]: http://radmind.cvs.sourceforge.net/viewvc/radmind/radmind/cksum.c
[watchedinstall]: http://bitbucket.org/ptone/watchedinstall/

Merry Christmas

Leave a reply

Thank you to the varied shop staff in London this past week for being without exception polite, enthusiastic and helpful when I asked for help. Especially the girl in [HMV Bond Street][hmv] who spelt [Zappa][zappa] as “zapper” – she was cute. It made tedious shopping joyful.

Let’s do it again next year! Love Dave

[zappa]: http://www.zappa.com/
[hmv]: http://hmv.com

Context managers

Leave a reply

I was re-writing the exellent [watchedinstall][watchedinstall] tool and needed to simplify a particularly gnarly chunk of code that required three sub-proceses to be started and then killed after invoking another process. It occurred to me I could make these into context managers.

Previously the code was something like…

start(program1)
try:
start(program2)
except:
stop(program1)
raise

try:
start(program3)
except:
stop(program2)
stop(program1)
raise

try:
mainprogram()
finally:
stop(program3)
stop(program2)
stop(program1)

Of course that could have been written with nested try / except / else / finally blocks as well, which I did start with but found not much shorter while almost incomprehensible.

[With context managers][ctxt] the whole thing was written as…

# from __future__ import with_statement, Python 2.5

with start(program1):
with start(program2):
with start(program3):
mainprogram()

So much more comprehensible! Here’s the implementation of the context manager (using the `contextlib.contextmanager` decorator for a triple word score):

import contextlib
import os
import signal
import subprocess

@contextlib.contextmanager
def start(program_args):
prog = subprocess.Popen(program_args)
if prog.poll(): # Anything other than None or 0 is BAD
raise subprocess.CalledProcessError(prog.returncode, program_args[0])

try:
yield
finally:
if prog.poll() is None:
os.kill(prog.pid, signal.SIGTERM)

For bonus points I might have used [`contexlib.nested()`][ctxtlib] to put the three `start()` calls on one line but then what would I do for the rest of the day?

[watchedinstall]: http://bitbucket.org/ptone/watchedinstall/
[ctxt]: http://docs.python.org/library/stdtypes.html#typecontextmanager
[ctxtlib]: http://docs.python.org/library/contextlib.html

Snow Leopard: a reactionary writes

Leave a reply

Things I like about [Mac OS X version 10.6][sl]:

(Mac OS X 10.6 is also known as Snow Leopard, although I dislike Apple’s use of the operating system codename in their publicity material because it leads to conversations where people talk about “Leopard” and “Tiger” and one has to stop for a second to translate those to actual operating system versions and no-one is ever going to refer to [Mac OS X 10.3 as Panther][panther] these days, let alone [10.2 being Jagwire][jaguar] or heaven forbid [Puma][puma] and [Cheetah][cheetah]. What are the chances I’ll have to look up the codename for 10.5 by the time we reach 10.10? Version numbers are not so evocative but are less confusing than codenames. This doesn’t mean I will stop naming hard disks after Mac OS codenames – my desktop has [Veronica, Gershwin, Harmony and Sonata][codenames] connected at the moment, with [Copland][copland] and [Pink][pink] sitting on the shelf as appropriate…)

Things I like about Mac OS X Snow Leopard:

– Apple’s drivers for my Epson all-in-one printer / scanner actually work. Epson’s drivers for the same printer / scanner only worked if you never used the scanner and promised to attend church more often.
– Significantly snappier.
– QuickTime Player’s minimal interface.

Things I dislike about Mac OS X Snow Leopard:

– By default the Finder does not show internal disks on the Desktop.
– The Finder [ignores type / creator codes][typecreator] on files.

Everything else in 10.6 is good. However it strikes me that the de-emphasizing of old-style Mac metadata (type / creator codes) and the default of not showing your computer’s hard drive icon on the desktop are evidence of the triumph of old-school Next-ies within Apple.

I think the decision to cover-up the hierarchical filesystem is a bad thing.

P.S. Wouldn’t it have been awesome if, having released Mac OS X Cheetah, Apple had continued with naming their releases after other famous Hollywood animal actors? Why they stopped naming releases after [disappointing Sylvester Stallone movies][sly] is beyond me – most any version of System 7 could have been named [Lock Up][lockup].

[sl]: http://www.apple.com/macosx/
[leopard]: http://www.apple.com/support/leopard/
[panther]: http://www.apple.com/support/panther/
[jaguar]: http://www.apple.com/support/jaguar/
[puma]: http://en.wikipedia.org/wiki/Mac_OS_X_v10.1
[cheetah]: http://en.wikipedia.org/wiki/Mac_OS_X_v10.0
[codenames]: http://www.mackido.com/CodeNames/MacOSSoftware.html
[copland]: http://lowendmac.com/orchard/05/1108.html
[pink]: http://lowendmac.com/orchard/05/1026.html
[lockup]: http://www.imdb.com/title/tt0097770/
[sly]: http://www.imdb.com/title/tt0118887/
[typecreator]: http://arstechnica.com/staff/fatbits/2009/09/metadata-madness.ars

Serving custom Django admin media in development

1 Reply

I’ve just discovered [Django][django]’s development server always serves admin media. This is tremendously useful because it means you don’t need to configure a static serve view in your project `urls.py` during development.

However what bit me was I wanted to use a customised set of admin media and had configured a view for the `ADMIN_MEDIA_URL` path and was going batty trying to work out why Django was ignoring it. It used to be that as long as you had `DEBUG = False` in `settings.py` then the development server did not try to help serve the admin media automatically.

[Changeset 6075][6075] added a switch to the runserver command for over-riding the admin media directory.

python manage.py runserver –adminmedia /path/to/custom/media

That change was made more than two years ago. It is [right there in the documentation][docs]. A little bit of magic that wasted fifteen minutes of my frantic schedule (except for the fact I do not have a frantic schedule).

[django]: http://www.djangoproject.com/
[6075]: http://code.djangoproject.com/changeset/6075
[docs]: http://docs.djangoproject.com/en/dev/ref/django-admin/#djadminopt—adminmedia

I am very bad at writing tests

Leave a reply

… but I _think_ I might be getting a little better.

At least these days when I am writing some script (almost certainly in [Python][python]) I start out by intending to write tests. I usually fail because I haven’t learnt to think in terms of writing code that can be easily tested.

[Mark Pilgrim][pilgrim]’s [Dive Into Python][dive] has great stuff on how to approach a problem by [defining the tests first and gradually filling in the code][divetest] that satisfies the test suite. One day I may be able to work like that, until then I work by writing a concise docstring, then stubbing out the function. Once the function is in a state where it might actually return a meaningful result I can play with it in the Python interpreter and start adding useful [doctests][doctest] to the [docstring][docstring].

What really helps is to break the logic out into tiny pieces where ideally each piece returns the result of transforming the input (which I think is known as a [functional approach][functional]). By doing this I can have tests for most of the code and those functions that have a lot of conditional logic, those functions that are harder to write tests for, will at least be relying on sub-routines that are themselves well tested.

I can dream.

[python]: http://www.python.org/
[pilgrim]: http://diveintomark.org/
[dive]: http://www.diveintopython.org/
[divetest]: http://diveintopython.org/unit_testing/stage_1.html
[doctest]: http://docs.python.org/library/doctest.html
[functional]: http://en.wikipedia.org/wiki/Functional_programming
[docstring]: http://www.python.org/dev/peps/pep-0257/

The hidden depths of Adobe CS4

Leave a reply

[Adobe][adobe]’s installers and updaters for the Creative Suite are amazingly bad. The updaters actually create a hidden directory `/Applications/.AdobePatchFiles` and store what I assume are the old versions of the files that get updated. Almost a gigabyte of data on my system!

What the fuck is wrong with [these guys][oobe]?

Not certain which is worse, that the updaters created a folder in `/Applications` that clearly belongs somewhere in `/Library/Application Support` (if it should exist at all) or that they made it hidden.

You can delete it.

[adobe]: http://www.adobe.com/
[oobe]: http://blogs.adobe.com/OOBE/

Crazy Acrobat installers love Python

Leave a reply

Looking through the updaters for [Adobe Acrobat][acrobat] 9 for Mac I came across a bunch of scripts written in [Python][python]. My favourte was called `FindAndKill.py`:

#!/usr/bin/python
“””
Search for and kill app.
“””
import os, sys
import commands
import signal

def main():
if len(sys.argv) != 2:
print ‘Missing or too many arguments.’
print ‘One argument and only one argument is required.’
print ‘Pass in the app name to find and kill (i.e. “Safari”).’
return 0

psCmd = ‘/bin/ps -x -c | grep ‘ + sys.argv[1]
st, output = commands.getstatusoutput( psCmd )

if st == 0:
appsToKill = output.split(‘\n’)
for app in appsToKill:
parts = app.split()
killCmd = ‘kill -s 15 ‘ + parts[0]
#print killCmd
os.system( killCmd )

if __name__ == “__main__”:
main()

(You can [download the Acrobat 9.1.3 update][acrobat913] and find this script at `Acrobat 9 Pro Patch.app/Contents/Resources/FindAndKill.py`.)

Was the author not aware of the `killall` command for sending a kill signal to a named process? The [`killall` man page][mankillall] says it appeared in [FreeBSD 2.1, which was released in November 1995][fbsd]. Adobe CS4 was [released about 14 years later][cs4]. How is it Adobe’s product managers approve these things for release?

What is particularly galling about Adobe’s Acrobat 9 updaters is that they seem to re-implement so much of what the Apple installer application does, even down to their use of gzipped cpio archives for the payload.

[acrobat913]: http://www.adobe.com/support/downloads/detail.jsp?ftpID=4538
[acrobat]: http://www.adobe.com/products/acrobatpro/
[python]: http://www.python.org
[mankillall]: http://www.manpagez.com/man/1/killall/
[fbsd]: http://www.freebsd.org/releases/2.1R/announce.html
[cs4]: http://www.adobe.com/aboutadobe/pressroom/pressreleases/200809/092308AdobeCS4Family.html

Migrating a Filemaker database to Django

3 Replies

At work we have several [Filemaker Pro][fmp] databases. I have been slowly working through these, converting them to Web-based applications using [the Django framework][django]. My primary motive is to replace an overly-complicated Filemaker setup running on four Macs with a single 2U rack-mounted server running [Apache][apache] on [FreeBSD][fbsd].

At some point in the process of re-writing each database for use with Django I have needed to convert all the records from Filemaker to Django. There exist good [Python][python] libraries for [talking to Filemaker][pyfmp] but they rely on the XML Web interface, meaning that you need Filemaker running and set to publish the database on the Web while you are running an import.

In my experience [Filemaker’s built-in XML publishing interface][fmpxml] is too slow when you want to migrate tens of thousands of records. During development of a Django-based application I find I frequently need to re-import the records as the new database schema evolves – doing this by communicating with Filemaker is tedious when you want to re-import the data several times a day.

So my approach has been to export the data from Filemaker as XML using [Filemaker’s FMPXMLRESULT][fmpxmlresult] format. The Filemaker databases at work are _old_ (Filemaker 5.5) and perhaps things have improved in more recent versions but Filemaker 5/6 is a very poor XML citizen. When using the FMPDSORESULT format (which has been dropped from more recent versions) it will happily generate invalid XML all over the shop. The FMPXMLRESULT format is better but even then it will emit invalid XML if the original data happens to contain funky characters.

So here is [filemaker.py, a Python module for parsing an XML file produced by exporting to FMPXMLRESULT][dave] format from Filemaker.

To use it you create a sub-class of the `FMPImporter` class and over-ride the `FMPImporter.import_node` method. This method is called for each row of data in the XML file and is passed an XML node instance for the row. You can convert that node to a more useful dictionary where keys are column names and values are the column values. You would then convert the data to your Django model object and save it.

A trivial example:

import filemaker

class MyImporter(filemaker.FMPImporter):
def import_node(self, node):
node_dict = self.format_node(node)
print node[‘RECORDID’], node_dict

importer = MyImporter(datefmt=’%d/%m/%Y’)
filemaker.importfile(‘/path/to/data.xml’, importer=importer)

The `FMPImporter.format_node` method converts values to an appropriate Python type according to the Filemaker column type. Filemaker’s `DATE` and `TIME` types are converted to Python [`datetime.date`][dtdate] and [`datetime.time`][dttime] instances respectively. `NUMBER` types are converted to Python `float` instances. Everything else is left as strings, but you can customize the conversion by over-riding the appropriate methods in your sub-class (see the source for the appropriate method names).

In the case of Filemaker `DATE` values you can pass the `datefmt` argument to your sub-class to specify the date format string. See Python’s [time.strptime documentation][strptime] for the complete list of the format specifiers.

The code uses [Python’s built-in SAX parser][pysax] so that it is efficent when importing huge XML files (the process uses a constant 15 megabytes for any size of data on my Mac running Python 2.5).

Fortunately I haven’t had to deal with Filemaker’s repeating fields so I have no idea how the code works on repeating fields. Please let me know if it works for you. Or not.

[Download filemaker.py][dave]. This code is released under a 2-clause BSD license.

[dave]: http://reliablybroken.com/b/wp-content/uploads/2009/11/filemaker.py
[strptime]: http://docs.python.org/library/time.html#time.strftime
[fmp]: http://www.filemaker.com/
[django]: http://www.djangoproject.com/
[apache]: http://httpd.apache.org/
[fbsd]: http://www.freebsd.org/
[python]: http://www.python.org/
[pyfmp]: http://code.google.com/p/pyfilemaker/
[fmpxml]: http://www.filemaker.com/support/technologies/xml
[fmpxmlresult]: http://www.filemaker.com/help/html/import_export.16.30.html#1029660
[dtdate]: http://docs.python.org/library/datetime.html#date-objects
[dttime]: http://docs.python.org/library/datetime.html#time-objects
[pysax]: http://docs.python.org/library/xml.sax.html

Network users and Mac 10.5 archive and install

Leave a reply

When upgrading a Mac from Mac OS X 10.4 (Tiger) to 10.5 (Leopard), remember that network accounts are _not_ included if you do an archive and install and choose to migrate existing users. If a network account had its home folder at `/Users/jbloggs` then it will have been moved to `/Previous Systems.localized/2009-11-06_0346/Users/jbloggs` (although the date portion will be the date that you did your install).

This applies to [network accounts which authenticate against Active Directory and do not have a mobile account][kb].

Why my place of work used to setup Macs with the option for create mobile account at login turned off is a mystery to me.

[kb]: http://docs.info.apple.com/article.html?path=ServerAdmin/10.5/en/c7od45.html