Python-ideas September 2013

python-ideas@python.org

67 participants
43 discussions

Specify number of items to allocate for array.array() constructor

by Sven Rahmann

At the moment, the array module of the standard library allows to create arrays of different numeric types and to initialize them from an iterable (eg, another array). What's missing is the possiblity to specify the final size of the array (number of items), especially for large arrays. I'm thinking of suffix arrays (a text indexing data structure) for large texts, eg the human genome and its reverse complement (about 6 billion characters from the alphabet ACGT). The suffix array is a long int array of the same size (8 bytes per number, so it occupies about 48 GB memory). At the moment I am extending an array in chunks of several million items at a time at a time, which is slow and not elegant. The function below also initializes each item in the array to a given value (0 by default). Is there a reason why there the array.array constructor does not allow to simply specify the number of items that should be allocated? (I do not really care about the contents.) Would this be a worthwhile addition to / modification of the array module? My suggestions is to modify array generation in such a way that you could pass an iterator (as now) as second argument, but if you pass a single integer value, it should be treated as the number of items to allocate. Here is my current workaround (which is slow): def filled_array(typecode, n, value=0, bsize=(1<<22)): """returns a new array with given typecode (eg, "l" for long int, as in the array module) with n entries, initialized to the given value (default 0) """ a = array.array(typecode, [value]*bsize) x = array.array(typecode) r = n while r >= bsize: x.extend(a) r -= bsize x.extend([value]*r) return x

7 months

Implicit string literal concatenation considered harmful?

by Guido van Rossum

I just spent a few minutes staring at a bug caused by a missing comma -- I got a mysterious argument count error because instead of foo('a', 'b') I had written foo('a' 'b'). This is a fairly common mistake, and IIRC at Google we even had a lint rule against this (there was also a Python dialect used for some specific purpose where this was explicitly forbidden). Now, with modern compiler technology, we can (and in fact do) evaluate compile-time string literal concatenation with the '+' operator, so there's really no reason to support 'a' 'b' any more. (The reason was always rather flimsy; I copied it from C but the reason why it's needed there doesn't really apply to Python, as it is mostly useful inside macros.) Would it be reasonable to start deprecating this and eventually remove it from the language? -- --Guido van Rossum (python.org/~guido)

2 years, 6 months

while conditional in list comprehension ??

by Wolfgang Maier

Dear all, I guess this is so obvious that someone must have suggested it before: in list comprehensions you can currently exclude items based on the if conditional, e.g.: [n for n in range(1,1000) if n % 4 == 0] Why not extend this filtering by allowing a while statement in addition to if, as in: [n for n in range(1,1000) while n < 400] Trivial effect, I agree, in this example since you could achieve the same by using range(1,400), but I hope you get the point. This intuitively understandable extension would provide a big speed-up for sorted lists where processing all the input is unnecessary. Consider this: some_names=["Adam", "Andrew", "Arthur", "Bob", "Caroline","Lancelot"] # a sorted list of names [n for n in some_names if n.startswith("A")] # certainly gives a list of all names starting with A, but . [n for n in some_names while n.startswith("A")] # would have saved two comparisons Best, Wolfgang

6 years, 7 months

A nice __repr__ for the ast.* classes?

by Haoyi Li

Wouldn't it be nice if this >>> import ast >>> print repr(ast.parse("(1 + 1)").body[0].value) <_ast.BinOp object at 0x0000000001E94B38> printed something more useful? >>> print repr(ast.parse("(1 + 1)").body[0].value) BinOp(left=Num(n=1), op=Add(), right=Num(n=1)) I've been doing some work on macropy <https://github.com/lihaoyi/macropy>, which uses the ast.* classes extensively, and it's annoying that we have to resort to dirty-tricks like monkey-patching the AST classes (for CPython 2.7) or even monkey-patching __builtin__.repr (to get it working on PyPy) just to get eval(repr(my_ast)) == my_ast to hold true. And a perfectly good solution already exists in the ast.dump() method, too! (It would also be nice if "==" did a structural comparison on the ast.* classes too, but that's a different issue). -Haoyi

6 years, 10 months

Gzip and zip extra field

by Serhiy Storchaka

Gzip files can contains an extra field [1] and some applications use this for extending gzip format. The current GzipFile implementation ignores this field on input and doesn't allow to create a new file with an extra field. ZIP file entries also can contains an extra field [2]. Currently it just saved as bytes in the `extra` attribute of ZipInfo. I propose to save an extra field for gzip file and provide structural access to subfields. f = gzip.GzipFile('somefile.gz', 'rb') f.extra_bytes # A raw extra field as bytes # iterating over all subfields for xid, data in f.extra_map.items(): ... # get Apollo file type information f.extra_map[b'AP'] # (or f.extra_map['AP']?) # creating gzip file with extra field f = gzip.GzipFile('somefile.gz', 'wb', extra=extrabytes) f = gzip.GzipFile('somefile.gz', 'wb', extra=[(b'AP', apollodata)]) f = gzip.GzipFile('somefile.gz', 'wb', extra={b'AP': apollodata}) # change Apollo file type information f.extra_map[b'AP'] = ... Issue #17681 [3] has preliminary patches. There is some open doubt about interface. Is not it over-engineered? Currently GzipFile supports seamless reading a sequence of separately compressed gzip files. Every such chunk can have own extra field (this is used in dictzip for example). It would be desirable to be able to read only until the end of current chunk in order not to miss an extra field. [1] http://www.gzip.org/format.txt [2] http://www.pkware.com/documents/casestudies/APPNOTE.TXT [3] http://bugs.python.org/issue17681

6 years, 10 months

Stdlib YAML evolution (Was: PEP 426, YAML in the stdlib and implementation discovery)

by anatoly techtonik

FWIW, I am +1 on for the ability to read YAML based configs Python without dependencies, but waiting for several years is hard. Maybe try an alternative data driven development process as opposed to traditional PEP based all-or-nothing style to speed up the process? It is possible to make users happy incrementally and keep development fun without sacrificing too much on the Zen side. If open source is about scratching your own itches, then the most effective way to implement a spec would be to allow people add support for their own flavor without disrupting works of the others. For some reason I think most people don't need full YAML speccy, especially if final implementation will be slow and heavy. So instead of: import yaml I'd start with something more generic and primitive: from datatrans import yamlish Where `datatrans` is data transformation framework taking care of usual text parsing (data partitioning), partition mapping (structure transformation) and conversion (binary to string etc.) trying to be as fast and lightweight as possible, bringing vast field for future optimizations on algorithmic level. `yamlish` is an implementation which is not vastly optimized (datatrans to yamlish is like RPython to PyPy) and can be easily extended to cover more YAML (hopefully). Therefore the name - it doesn't pretend to parse YAML - it parses some supported subset, which is improved over time by different parties (if datatrans is done right to provide readable (maintainable + extensible) code for implementation). There is an exisiting package called `yamlish` on PyPI - I am not talking about it - it is PyYAML based, which is not an option for now as I see it. So I stole its name. Sorry. This PyPI package was used to parse TAP format, which is again, a subset. Subset.. It appears that YAML is good for humans for its subsets. It leaves an impression (maybe it's just an illusion) that development work for subset support can also be partitioned. If `datatrans` "done right" is possible, it will allow incremental addition of new YAML features as the need for them arises (new data examples are added). Or it can help to build parsers for YAML subsets that are intentionally limited to make them performance efficient. Because `datatrans` is a package isolating parsing, mapping and conversion parts of the process to make it modular and extensible, it can serve as a reference point for various kinds of (scientific) papers including the ones that prove that such data transformation framework is impossible. As for `yamlish` submodule, the first important paper covering it will be a reference table matrix of supported features. While it all sounds too complicated, driving development by data and real short term user needs (as opposed to designing everything upfront) will make the process more attractive. In data driven development, there are not many things that can break - you either continue parsing previous data or not. The output from the parsing process may change over time, but it may be controlled by configuring the last step of data transformation phase. `Parsing AppEngine config file` or `reading package meta data` are good starting points. Once package meta data subset is parsed, it is done and won't break. The implementation for meta data parsing may mature in distutils package, for AppEngine in its SDK, and merged in either of those, sending patches for `datastrans` to stdlib. The question is only to design output format for the parse stage. I am not sure everything should be convertible into Python objects using the "best fit" technique. I will be pretty comfortable if target format will not be native Python objects at all. More than that - I will even insist to avoid converting to native Python object from the start. The ideal output for the first version should be generic tree structure with defined names for YAML elements. The tree that can be represented as XML where these names are tags. The tree can be therefore traversed and selected using XPath/JQuery syntax. It will take several years for implementation to mature and in the end there will be a plenty of backward compatibility matters with the API, formatting and serializing. So the first thing I'd do is [abandon serialization]. From the point of view of my self-proclaimed data transformation theory, the input and output formats are data. If output format is not human readable - as some linked Python data structures in memory - it wastes time and hinders development. Serializing Python is a problem of different level, which is an example of binary, abstract, memory-only output format - a lot of properties that you don't want to deal with while working with data. To summarize: 1. full spec support is no goal 2. data driven (using real world examples/stories) 3. incremental (one example/story at a time) 4. research based (beautiful ideas vs ugly real world limitations) 5. maintainable (code is easy to read and understand the structure) 6. extensible (easy to find out the place to be modified) 7. core "generic tree" data structure as an intermediate format and "yaml tree" data structure as a final format from parsing process P.S. I am willing to wok on this "data transformation theory" stuff and prototype implementation, because it is generally useful in many areas. But I need support. -- anatoly t.

6 years, 10 months

'from os.path import FILE, DIR' or internal structure of filenames

by anatoly techtonik

FILE = os.path.abspath(__file__) DIR = os.path.abspath(os.path.dirname(__file__)) ? Repeated pattern for referencing resources relative to your scripts. Ideas about alternative names / locations are welcome. In PHP these are __FILE__ and __DIR__. For Python 3 adding __dir__ is impossible, because the name clashes with __dir__ method (which is not implemented for module object, but should be [ ] for consistency). Also current __file__ is rarely absolute path, because it is never normalized [ ]. So it will be nice to see normalization of Python file name after the import to reduce mess and make its behaviour predictable - http://stackoverflow.com/questions/7116889/python-file-attribute-absolute-o… ----[ possible spec. draft for a beautiful internal structure ]-- The Python interpreter should provide run-time information about: 1. order of import sequence 2. names of imported modules 3. unique location for each imported module which unambiguously identifies it 4. run-time import dependency tree (not sure about this, but it can help with debugging) 5. information about sys.path entry where this module was imported from 6. information about who and when added this sys.path entry -- anatoly t.

6 years, 11 months

Python 3.4 should include docopt as-is

by anatoly techtonik

This - http://docopt.org/ - should be included with Python 3.4 distribution. -- anatoly t.

6 years, 11 months

PyPi per-file download counters

by Giampaolo Rodola'

I recently moved psutil .tar.gz and .exe files from Google Code to PyPi and noticed it doesn't show total per-file download counters: https://pypi.python.org/pypi?:action=display&name=psutil#downloads Why don't we add them? Thoughts? --- Giampaolo http://code.google.com/p/pyftpdlib/ http://code.google.com/p/psutil/ http://code.google.com/p/pysendfile/

6 years, 11 months

An exhaust() function for iterators

by Clay Sweetser

Currently, several strategies exist for exhausting an iterable when one does not care about what the iterable returns (such as when one merely wants a side effect of the iteration process). One can either use an empty for loop: for x in side_effect_iterable: pass A throwaway list comprehension: [x for x in side_effect_iterable] A try/except and a while: next = side_effect_iterable.next try: while True: next() except StopIteration: pass Or a number of other methods. The question is, which one is the fastest? Which one is the most memory efficient? Though these are all obvious methods, none of them are both the fastest and the most memory efficient (though the for/pass method comes close). As it turns out, the fastest and most efficient method available in the standard library is collections.deque's __init__ and extend methods. from collections import deque exhaust_iterable = deque(maxlen=0).extend exhaust_iterable(side_effect_iterable) When a deque object is initialized with a max length of zero or less, a special function, consume_iterator, is used instead of the regular element insertion calls. This function, found at http://hg.python.org/cpython/file/tip/Modules/_collectionsmodule.c#l278, merely iterates through the iterator, without doing any work allocating the object to the deque's internal structure. I would like to propose that this function, or one very similar to it, be added to the standard library, either in the itertools module, or the standard namespace. If nothing else, doing so would at least give a single *obvious* way to exhaust an iterator, instead of the several miscellaneous methods available. -- "Evil begins when you begin to treat people as things." - Terry Pratchett

6 years, 11 months

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

Python-ideas September 2013