🌐 Mailman 3 February 2002 - Tutor - python.org

Creating typed files with Python, BeOS (OT)
by Charlie Clark May 12, 2002

May 12, 2002

>Dear Python people, > >How can you tell Python to make a simple text file or to make a kind of file >with your own kind of extension? I have looked around in books I have, and >online. I have only found stuff about how to open and close and recieve >input from. I don't know enough about file creation and operating systems >to know I guess how to tell Python to do it. Creating typed files depends on the operating system's support for file types. Extensions are not the same as file types!!! Python will let you create files with any name as long as it is allowd by the operating system: f = open("ourfile", "w") Giving textfiles a ".txt" extension is purely a convention required for primitive file systems such as FAT/ NTFS or Posix. Removing or changing the extension can make such files unreadable on that system. Some file systems such as the BFS provide sophisticated file typing and Python can hook into this. Bear in mind that this is platform specific The following code creates an e-mail file in BeOS from BeOS.fsattr import write_attr newfile = "newfile.mail" # creat an empty file f = open(newfile, "w") f.close() #give our new file a file type write_attr(newfile, "BEOS:TYPE", B_MIME_STRING_TYPE, "text/x-email\0", 0) Our file "newfile.mail" now has the MIME-type "text/x-email". E-mail files are a subset of textfiles and can be read by any text editor or mail program independent of the file extension. Charlie

3 2

Glade
by djuro m. May 7, 2002

May 7, 2002

Hello everyone! Could someone please "explain" to me how to make a simple gui working with python code by using Glade. ( I only need to see a code or how is python code tied with gui code file) I already tried several times according to instructions in available "Getting started-s" but it didn't work. I use RedHat 7.0, Python 1.5.2 and Glade 0.5.9 For example by Tkinter: from Tkinter import * top = Tk() e = Entry() e.pack() def Write(): global e m = "Hi there!" e.insert(END,m) b = Button(text='OK',command = Write) b.pack() top.mainloop() Thank you D. M.

6 5

A small math puzzle [recreational Python]
by Danny Yoo May 2, 2002

May 2, 2002

Hi everyone, I ran across this puzzle while browsing the the 'alt.math.undergrad' newsgroup. I thought it might make a fun programming exercise, so here's my paraphrase of the puzzle: A "rearrangement without fixed letters" of a given word is a mixup of the letters so that, if one were to overlap the original with the rearrangement, no letters should match. For example, given the word: 'TERESA' here are a two "rearrangements without fixed letters": ['RTESAE', 'SATREE'] However, here's an example that is a rearrangement but doesn't work because it has a fixed letter: 'TRESAE' ^ In this example, the 'T' makes this rearrangement inacceptable, since 'TERESA' itself begins with a 'T' too. The puzzle is to write a program that counts all the possible rearrangements without fixed letters. Have fun!

8 16

Re: Print ogre
by Charlie Clark March 1, 2002

March 1, 2002

>| Where else does ">>" occur in Python? > >Right shift. .. >It is a bitwise operator. aa is 10101010 in binary. Right shift it >one place and you get 01010101 which in hex is 55. Shift it two >places and you get 00101010 which is 2a in hex. In other words there is no overlap between the two uses. ">>" doesn't have a generic meaning or use making reusable in different contexts. A good reason for removing it in print >> as a way to remove a possible source of confusion. Time to check the reasons for introducing it and the rules for writing a PEP. Charlie

8 13

Re: [Tutor] types
by clickron＠webtv.net Feb. 28, 2002

Feb. 28, 2002

That's what I was looking for. Thanks for the help. I'm learning even though it's not happening quickly. But I'm having fun. Ron

1 0

packet parsing
by python＠jayed.com Feb. 28, 2002

Feb. 28, 2002

Hello, I'm trying to make the best of my unemployed time and have decided that implementing an SSH client in Python will keep me occupied for a few years. I've just started, and am working on the first non-trivial part (which appeared REALLY quickly in the IETF draft). My program connects to the server; the server says, "I'm an SSH server". My client says, "I'm an SSH client". The server then sends a variable length message in SSH Binary Packet Format to initiate a key exchange. My first question is: do I need to be concerned about little/big/network-endian? Both of my machines are little-endian. So when I look at a byte from the SSH server's packet, do I need to do a network-to-host translation? Conversely, when I send a byte to the server, do I need to do a host-to-network translation My second question is: how do I look at/manipulate bytes of an incoming network packet? Or bits even. Right now, I have the following piece of code that receives the SSH server's initial key exchange packet: kexrecv = mysock.recv(blocksize) I want to directly manipulate the bits/bytes of kexrecv. Everything that I've done makes kexrecv into a string. And I'm having problems with kexrecv as a string -- I want to deal with it as a binary stream. (Note: this might not be the best way to deal with it, but until I have a good grip on the SSH transport protocol, it's how I prefer to deal with it -- right now I want to be able to say: "read the first 32 bits, convert it into a number; read the next 8 bits, convert it into a number; read the next 128 bits...blah, blah, etc.). type(kexrecv) results in <type 'str'> type(`kexrecv`) results in <type 'str'> len(kexrecv) returns nothing len(`kexrecv`) returns 830 print len(kexrecv) returns 632 print len(`kexrecv`) returns 830 (Yes, I know that the backticks are equivalent to repr() -- I looked it up on google last night). I want to deal with kexrecv and not `kexrecv`. But I haven't been able to figure it out. If kexrecv is a string [as type(kexrecv) seems to think] why doesn't len(kexrecv) return anything? And if kexrecv is a string, why does len(`kexrecv`) return a larger value than "print len(kexrecv)"? Thanks, Jay

3 4

how to get complete output in interpreter?
by Christian Ebert Feb. 28, 2002

Feb. 28, 2002

Hello there, Is there an upper limit to the (lines of?) print statements in the interpreter? As I don't want to bother with a gui yet I use the interpreter as interface for a little "database". Everything works fine, except when I print a "long" result of a query. The interpreter then truncates the result at the top and displays only a certain amount of lines. It's not due to an error in the script because when I write the result to a file it's all there. Splitting into several print statements doesn't make any difference either. Anybody knows how to change this? Or is it a MacPython issue (I will ask on the macpython list then)? TIA Christian -- Hyperion in Frankfurt www.tko.ch/blacktrash www.schauspielfrankfurt.de

2 1

types
by clickron＠webtv.net Feb. 28, 2002

Feb. 28, 2002

I want to ask for a integer using something like number = input("Give me a number'), but if a string is accidentally put in I don't want it to throw an error message. I'd like it to say something like "Enter a number, try again." I'm sure it's something simple and I'm just not seeing it. Appreciate any help out there. Ron

3 2

RE: [Tutor] Problems with genetically engineering the Print Ogre
by alan.gauld＠bt.com Feb. 28, 2002

Feb. 28, 2002

> In America, students are exposed to "set builder" notation > starting at about age 12. That's where list comprehensions > trace from, quite directly I can't speak for Germany, nor even for modern UK practice, but personally I didn't meet set definition syntax until the end of second year high school (about 14) and had the opportunity to drop math (and major on say sciences/engineering instead) very soon after. Thus it is entirely possible that I could have gone thru' school with only a brief exposure to set building notation. (We had of course covered set theory with Venn diagrams, Union symbols etc but not the set definition syntax resembling comprehensions...) In fact I personally kept up math and thus had exposure through the next 3 years until I went off to earn a living! Then, 9 years after leaving school I went back to University and studied Electrical Engineering with compulsory math each year for 4 years. I think I only saw set notation being used a handful of times and it certainly wasn't needed to pass the exams. My point being that in my own experience it is very easy to become a professional engineer/programmer without much exposure to set builder notation and in that case list comprehensions are far from being easily comprehensible! As for me, I think they are overused - like many new toys - and the older map/filter/reduce are often clearer constructs, but in the other places, where they really are useful I am starting to use them and grudgingly like them.... I just don't find the [x for x ... syntax scans well in my brain... > You can't tell me Germany[or UK?] coddles its young > students more than America does <wink>. It may be so. Or maybe we give them integral calculus sooner to compensate - around the same time as set builder notation... :-) Alan G.

1 0

RE: [Tutor] Problems with genetically engineering the Print Ogre
by alan.gauld＠bt.com Feb. 28, 2002

Feb. 28, 2002

> Since I came into the discussion late - did anybody ever > suggest using a format like > > to <file> print <stuff> > > or, as not to use up "to" as a word, > > into <file> print <stuff> This is just a 'natural' English translation of OO syntax, thus: file.print() It clutters up the language to little advantage IMHO. If file objects got a method that acted like the print statement (foibles and all) then that would IMHO suffice. BTW I'm astounded this dsiscussion is still going on from when I lost contact over a week ago! :-) Alan g.

1 0