Is mathematica.stackexchange.com/questions/67388/… still a bug in 10.1?

@hftf It seems fine on 10.1 Windows. I normally have the suggestions bar turned off, but I turned it on and tried the test sequence and AssociationMap worked.

Updated category browser to 10.1. Try it out: en.wikipedia.org/wiki/User:Wakebrdkid/Popular_category_browsing

@Kuba you have a answer about package load structure don't you? Do you know where it is? It's about init.m package load structure.

@Murta Hello :) this one? How to use Get on a folder

@Kuba Yes! Tks. I'm with some problems to read a Package in V10.1 that I haven't in V10.0.2

@Murta when MMa is supposed to do something strange automatically I usually put explicit code/path. I don't have time to debug not mine mistakes :P

@Kuba Do we really want the bugs tag for documentation errors? There's plenty of problems with the documentation and they're fixed separately from Mathematica the software. People looking at the question and seeing the bugs tag are going to think that this is incorrect behaviour. It's not, it's just not clearly explained in the docs that the backtick is needed.

@Szabolcs I partially agree, but it is quite tough to say. Maybe in this question is not really necessary but in general it is. If a function gives wrong result in edge case because someone forgot to add "Possible issues" section in docs, is this acceptable?

@Szabolcs Or maybe different tag is an option, I do not insist on bug_documentation combination.

I think tagging everyhitng like this as "bug" just opens Pandora's gate and lots of unnecessary complaints about "how things should be". A bug is behaviour not intended by the developers. This is not that.

Also, I don't see much use for it. What is there to be gained from documenting and categorizing documentation mistakes?

@Szabolcs it is easier to find Get+bugs than Get+"path confusion" or whatever, the more that search engine is not perfect.

@Szabolcs p.s. maybe possible-issues or edge-cases? What do you think?

@Kuba I founded what changed in V10.1 that broked my code. I Was calling my packages with .wl, and It behavies just like .m files. But now that changed!

With .wl, if you do a Get in a directory like Get["dir`pack"], Get looks first for wl files, then for init.m, then form .m files...

In 10.0.2 it was first init.m files and then .wl and .m

@Murta Good to know

@Kuba Sorry for having left in a rush, I will be able to respond to everything later today ... just wanted to let you know I didn't disappear :)

@Szabolcs Sure, see you later :)

Isn't it a day early for stackegg, or is the real April fool's joke that March only has 30 days?

Hmmh... how do I interpolate a distribution I assume to continuous, but possibly nondifferentiable at integer values of x, and could be approximated with polynomials between integer value of x and the next one?

I suppose... mathematica.stackexchange.com/questions/56023/…

I downloaded and installed 10.1 at the office without much difficulties. However, on my home PC I get:

Anyone else got this?

virustotal doesn't report any problems, so I'll assume it's a false alarm.

A manual scan with my virus scanner (Avira) also doesn't reveal anything.

Hi @Murta

@SjoerdC.deVries I didn't have any problem.

@murta You can't be too cautious in these kinds of things

I pressed cancel in the above dialog box. It's "repairing system" (huh?) now, but it's hanging at 95%

@SjoerdC.deVries This happens sometimes. My reccomendation is to contact Wolfram Tech Support and let them know which virus scanner is doing this. They can get in contact with the company and let them know about the false flag (which it likely is)

To verify that your download is actually okay, the user portal has and MD5 checksum that you can use

@Searke Wouldn't that only make sense if I suspected that somebody intercepted my download? I mean, if the installer package would include a virus, that would end up as being part of the checksum, and won't tell anything.

huh... or at least it used to have md5 checksums on the user portal....

@Searke I know, but my point is that that isn't much of a virus check

@SjoerdC.deVries I don't understand. How would someone insert a virus into the download without changing the MD5 sum given by Wolfram Research?

Unless you are worried that Wolfram Research is distributing Mathematica with a virus in it already

@Searke If the guy who finalized the installer package was contaminated, the virus would enter the package and the MD5 would include that too.

Not the first time it happens

Anyway, I don't think we have that here

Well. I have absolutely no means of disproving you

but I've seen anti virus programs flag mathematica installers several times

usually we have to talk to the anti virus company and basically have to file a bug report with them.

@Searke Virustotal.com has examined it with like 30 scanners. None of them reported trouble

though the first time the avira scanner timed out.

I ran the whole stuff again and this time all was clear.

At any rate, if you're still worried, tech support should be able to give you an MD5 sum that would rule out some sort of man in the middle swap-out

although they don't do this commonly and I'm sure they're swamped since they always are during a release.

If you'd also like to report the fact that it your anti-virus does this, they have been able to handle that in the past as well.

@kirma Yes. GrammarRules is aimed at natural language.

GrammarTokens are very similar to Interpreter. They take a natural language description of something and give back a concrete representation of it that you can do computations with.

Can anyone review this question and answer? I'm hoping to be proven wrong. I've run into this problem before...

@Szabolcs you are probably right. Btw, you can use ToExpression with those methods: mathematica.stackexchange.com/q/73172/5478

or not exactly. Please skip second sentence :)

how about something like ToExpression["1.23", StandardForm, HoldComplete]

still no error checking

@chuy and then MatchQ[..., HoldComplete[_Integer | _Real]] ? Sounds like a good idea!