🌐 Mailman 3 February 2024 - Python-list - python.org

Re: Matplotlib warning [error?] message
by Zahraa Fadhil Feb. 18, 2024

Feb. 18, 2024

On Sunday, February 18, 2024 at 10:48:29 PM UTC+3, Leif Svalgaard wrote: > The latest[?] version of Matplotlib cannot show a figure. I get the > annoying error message: "Matplotlib is currently using agg, which is a > non-GUI backend, so cannot show the figure" > I'm using Spyder python 3.11 on Windows 11. > What to do? > > -- > Leif Svalgaard > I have the same problem try this 1) add this line first %matplotlib inline import matplotlib import matplotlib.pyplot as plt 2) delete this line from your cod.....matplotlib.use('Agg') 3) save the result into png file plt.savefig('D:\data\mygraph.png') another solution https://www.w3schools.com/python/trypandas.asp?filename=demo_ml_dtree4 import sys import matplotlib matplotlib.use('Agg') .... import matplotlib.pyplot as plt .... #Two lines to make compiler able to draw: plt.savefig(sys.stdout.buffer) sys.stdout.flush()

1 0

Matplotlib warning [error?] message
by Leif Svalgaard Feb. 18, 2024

Feb. 18, 2024

The latest[?] version of Matplotlib cannot show a figure. I get the annoying error message: "Matplotlib is currently using agg, which is a non-GUI backend, so cannot show the figure" I'm using Spyder python 3.11 on Windows 11. What to do? -- Leif Svalgaard leif(a)leif.org

1 0

Re: Using __new__
by dn Feb. 18, 2024

Feb. 18, 2024

On 18/02/24 13:21, Jonathan Gossage wrote: > The problem is that if you are dealing with a library class, you may > have times when the superclass is 'object' while at other times, with a > different inheritance hierarchy, the superclass may need arguments. My > thought is that the object class __new__ method should not choke on > arguments, just ignore them. All true. So, what you're looking for is one mechanism to rule them all? Not going to happen: for exactly the reasons you've stated. If you really want to get right 'down into the weeds' with a __new__() constructor, then you're well into customisation-territory. I think it would be 'going nuts' but... If it 'absolutely, positively, ...' then perhaps introspect the super-class and modify the call based-upon whether it is 'something' or "object"? (in similar fashion to the singleton's if-statement attempting to make sure it is unique) - perhaps someone knows a better/proper way to do this? Suggested research: custom classes, ABCs, and meta-classes... See also recent improvements to Python which have made it easier for sub-classes (and Descriptors - __set_name__() ) to identify who/how/where to 'phone home', in case (also) applicable... > When I talk about 'object', I am talking about the ultimate base class > of any inheritance hierarchy. have seen the class named 'object' called > that. Correct. The earlier comment was that class S( object ): is 'tradition', and synonymous with: class S: (not disputing the concept of "object" as the base class) Not correct. Please see last paragraph from previous message: > On Sat, Feb 17, 2024 at 7:06 PM dn via Python-list > <python-list(a)python.org <mailto:python-list@python.org>> wrote: ... > PS please reply to the list - there may be others who can learn > from, or > contribute to, this conversation! ... -- Regards, =dn

2 1

Proposed Python Computer Program - Feb. 17, 2024
by Science Researcher Feb. 18, 2024

Feb. 18, 2024

PROPOSED PYTHON COMPUTER LANGUAGE PROGRAM - Posted on February 17, 2024 TOPICS Some Background Information Test Post Newsgroups Adding Posting Dates To Newsgroup Notes E-mail Address Other Internet Security Steps Personal Opinion Statements SOME BACKGROUND INFORMATION A fair amount of detail has been included with this Newsgroup post because now that I finally have my Newsgroup posting program running again, I am planning to soon start recommending that Python programmers around the world consider working on the creation of a very important python computer language program that I believe will be of great value to international disaster management efforts and to quite a few scientists around the world. The overall programming effort might best be run by university and government python programming personnel. But the computer program would have so many different uses that individual programmers would likely be able to make improvements and submit them for consideration by the university and government groups. A perl computer language version of the program has been running for years. A long time ago, code for an early version was formally copyrighted in the United States. Program code was also made available for free use under a GNU license. The version of perl that I am using processes computer program code in a very slow manner. The present perl program contains more than 13,000 line of code and has to generate png picture files by sending information to a gnuplot program. Compiled python programs would run much faster and be of greater value to scientists. Gnuplot, for example, would no longer be needed. I expect that once a python version of the program became available to scientists, the program would evolve with tremendous speed. It should have the ability to tell scientists how to make improvements to its own math routines. They would improve the program's data generation accuracy and its value. One of my specialty areas in science involves disaster management where I believe I am considered to be an "Expert." I have several degrees in the physical sciences. I do a fair amount of computer programming but would not be considered a python computer programming "Expert" or even a "Novice." So it would be difficult for me to translate the present perl and gnuplot routines to python. In the past, one programming problem resulted from the fact that I developed the original perl program with the help of a retired U.S. government computer programmer. He could work with gnuplot and TrueBasic but was unable to work with either perl or python. A number of compromises had to be made. One involved using gnuplot to generate picture files. TEST POST NEWSGROUPS Several response posts to my original test posts discussed sending them to some test Newsgroup before posting them to the comp.lang.python Newsgroup. I did post one or more of my original tests posts to the test.ignore Newsgroup. There were two problems that I was having was with my E-mail computer program - Windows Live Mail - that also creates my Newsgroup posts. First - It worked fine for E-mail after I installed it on my computer. But it would not work for Newsgroup posts. I called my E-mail server personnel about the problem. But they seemed to have no idea regarding what I was talking about and could not answer my questions. Second - my Newsgroup post program would not use the return E-mail address that I wanted it to use when it created Newsgroup posts. Solutions - The solutions for both problems were to just reset everything with my E-mail server and my E-mail and Newsgroup message creation program. Then the Newsgroup posting part of "Windows Live Mail" started working. ADDING POSTING DATES TO NEWSGROUP NOTES The contents of my Newsgroup posts are at times sent to people by E-mail or are posted to Internet forums. So, I usually add the posting date to my Newsgroup posts so that people receiving the E-mails or reading the forum posts know when the original Newsgroup message was posted. The E-mails etc. would just contain text information. People reading them would not be able to determine from the header information when the original Newsgroup note was posted. E-MAIL ADDRESS The response E-mail address attached to this Newsgroup post is a real E-mail address. It has a special form so that the address can be easily changed when spam robots start sending E-mails to that address. Over the years I have found that spam robots can create a variety of problems once they get hold of an E-mail address. Making a minor change in the E-mail address seems to take care of the problem. It appears that even a simple change can confuse most or all of the spam robots. For a while, they likely continue to send E-mails to an address that no longer exists and probably eventually abandon their efforts. OTHER INTERNET SECURITY STEPS Over the decades, I have sent a considerable amount of information to various world governments. They have at times made good use of my recommendations. With each election, people running different governments, especially the U.S. government, can change their opinions regarding what they believe is and is not important. As a consequence, people running the U.S. government, for example, might presently like what a person states in a Newsgroup post. But 4 or 8 years from now, a new group of people running the government might not like what someone said 4 or 8 years in the past. To be on the safe side, since various governments might be using my information regarding subjects such as "Renewable Energy," I don't add personal names to my Newsgroup posts or even to my E-mail messages. Instead, some neutral name such as "Science Researcher" is used. Most government officials would probably not be able to easily determine who "Science Researcher," for example, is. Except in unusual circumstances, they likely won't attempt to check the header information associated with a Newsgroup post or an E-mail letter. If you add a personal name to a Newsgroup post, an E-mail, an Internet forum post, or your own Web site you are making it easy for possibly unfriendly future government officials to determine whom you might be and then get revenge on you for proposing that "Renewable Energy" is good! PERSONAL OPINION STATEMENTS "These are statements of personal opinion." It is likely important to at times add the above "Personal Opinion" statement to Newsgroup posts when they contain specific words, phrases, or information etc. such as "Windows Live Mail" that might be copyrighted. Most people would probably not want to have the Microsoft corporation getting angry with them. I also don't want my E-mail server personnel to become angry because I stated that they were unable to answer some simple questions regarding posting notes to Newsgroups. Regards to all. Science Researcher

1 1

Re: test-ignore
by Science Researcher Feb. 18, 2024

Feb. 18, 2024

"Lawrence D'Oliveiro" wrote in message news:uqrik4$lrc1$1@dont-email.me... On Sat, 17 Feb 2024 17:00:59 -0600, Science Researcher wrote: > "Lawrence D'Oliveiro" wrote in message > news:uqmbp3$3hsa6$1@dont-email.me... > > If I remember correctly, I had to get the installation program > from a third party site. I’m sure that was a perfectly trustworthy site, no doubt ... Comment: My moderately expensive computer virus program can be instructed to check the contents of downloaded EXE programs and other types of programs or ZIP files. It appears to do a fairly good job with that. But, you are correct. People should be careful regarding what they download.

1 0

Testing (sorry)
by Grant Edwards Feb. 18, 2024

Feb. 18, 2024

Today I noticed that nothing I've posted to python-list in the past 3 weeks has shown up on the list. I don't know how to troubleshoot this other than by sending test messages. Obviously, if this shows up on the list, then I'm making progress... [message 3] -- Grant

1 0

Testing - 2 (sorry)
by Grant Edwards Feb. 18, 2024

Feb. 18, 2024

Today I noticed that nothing I've posted to python-list in past 3 weeks has shown up on the list. I don't know how to troubleshoot this other than sending test messages. Obviously, if this shows up on the list, then I've gotten it to work... -- Grant

1 0

Re: Using __new__
by dn Feb. 18, 2024

Feb. 18, 2024

On 18/02/24 12:48, Jonathan Gossage wrote: > The problem that I am facing is that when the superclass is not > 'object', the __init__ method may well need arguments. I do not know how > to determine if the superclass is 'object'. For what it is worth, any > attempt to use this with different arguments should return the initial > singleton and ignore further attempts to create a second instance. 1 "object" don't understand. Perhaps give us a broader description of the problem? Remember also ABCs (Abstract Base Classes). 2 arguments yes, must accommodate arguments in __new__() if some/same are needed in __init__() However, when using the default "object", the super() does not need, use, or want, any arguments to be passed. 3 singleton don't think that is happening! PS please reply to the list - there may be others who can learn from, or contribute to, this conversation! -- Regards, =dn

1 0

Can one output something other than 'nan' for not a number values?
by Grant Edwards Feb. 17, 2024

Feb. 17, 2024

[Posts via slrn and my GMail account aren't showing up, so I guess I'll try subscribing from a different e-mail address.] On 2024-02-17, Cameron Simpson via Python-list <python-list(a)python.org> wrote: > On 16Feb2024 22:12, Chris Green <cl(a)isbd.net> wrote: >> I'm looking for a simple way to make NaN values output as something >> like '-' or even just a space instead of the string 'nan'. [...] >> >> Battery Voltages and Currents >> Leisure Battery - 12.42 volts -0.52 Amps >> Starter Battery - 12.34 volts -0.01 Amps > The simplest thing is probably just a function writing it how you > want it: > > def float_s(f): > if isnan(f): > return "-" > return str(f) Since he's obviously using one of the float formatting mechanisms to control the number of columsn and decimal places, I doubt str(f) will meet the need. I tried monkey-patching the float type's __format__ method, but it's immutable. Is float.__format__() what's used by f-strings, the '%' operator, etc.? -- Grant

1 0

Can one output something other than 'nan' for not a number values?
by Grant Edwards Feb. 17, 2024

Feb. 17, 2024

[I've been trying all afternoon to post via slrn, but nothing is showing up on the list. Forgive me if multiple posts eventually show up.] On 2024-02-17, Cameron Simpson via Python-list <python-list(a)python.org> wrote: > On 16Feb2024 22:12, Chris Green <cl(a)isbd.net> wrote: >>I'm looking for a simple way to make NaN values output as something >>like '-' or even just a space instead of the string 'nan'. [...] >> >> Battery Voltages and Currents >> Leisure Battery - 12.42 volts -0.52 Amps >> Starter Battery - 12.34 volts -0.01 Amps > The simplest thing is probably just a function writing it how you > want it: > > def float_s(f): > if isnan(f): > return "-" > return str(f) Since he's obviously using one of the float formatting mechanisms to control the number of columsn and decimal places, I doubt str(f) will meet the need. I tried monkey-patching the float type's __format__ method, but it's immutable. Is float.__format__() what's used by f-strings, the '%' operator, etc.? -- Grant

1 0