Inclusion of worktodo.add file?
 

In other threads, there is the suggestion that it takes some time
for the worktodo.add file to be merged into the worktodo.txt file.
How much time does this normally take? I created a worktodo.add
file with a P-1 factoring job yesterday evening and, after almost
24 hours, it still hasn't been incorporated into the worktodo.txt file.
In that period, there have been a number of automatic connections
to the V5 server to report new results and get new exponents. I even
did a couple of manual communications of new completion dates. All
to no avail.

I hadn't the patience to wait as long as you did. So I stoppd P95, and restarted it. After that, my worktodo.add was migrated in the worktodo.txt file and deleted.....

[QUOTE=Phantomas;147276]I hadn't the patience to wait as long as you did. So I stoppd P95, and restarted it. After that, my worktodo.add was migrated in the worktodo.txt file and deleted.....[/QUOTE]

Already tried that, twice. No effect. Also, a reboot of the machine
had no effect.

Just looked more closely at the properties of the worktodo.add file
and I see that the filename is shown as "worktodo.add.txt". That might
be the reason it doesn't get incorporated. So I try renaming the file: the
menu rename doesn't display the ".txt". Load the file back into Notepad
and attempt to save it out under a different name: it seems
that Notepad automatically appends .txt for a text file. Wordpad was
no help either in removing the ".txt". Loaded file into swriter, unclicked
the "add automatic extension" and saved it under the desired name.
Stopped and restarted Prime95. Finally, .add file merged with .txt file.

Force manual communication. Reports "unsupported assignment work
type".

I really hate Windoze...

[quote=Graff;147282]Already tried that, twice. No effect. Also, a reboot of the machine
had no effect.

Just looked more closely at the properties of the worktodo.add file
and I see that the filename is shown as "worktodo.add.txt". That might
be the reason it doesn't get incorporated. So I try renaming the file: the
menu rename doesn't display the ".txt". Load the file back into Notepad
and attempt to save it out under a different name: it seems
that Notepad automatically appends .txt for a text file. Wordpad was
no help either in removing the ".txt". Loaded file into swriter, unclicked
the "add automatic extension" and saved it under the desired name.
Stopped and restarted Prime95. Finally, .add file merged with .txt file.

Force manual communication. Reports "unsupported assignment work
type".

I really hate Windoze...[/quote]
Ah, the classic problem of the extra ".txt". Here's a little tip for fixing that in Notepad: enclose the file name in quotes in the Save File dialog box. For example, to save a file as worktodo.add, type "worktodo.add" (including the quotes) in the file name box in the Save File window. :smile:

[quote=Graff;147282]
I really hate Windoze...[/quote]

There is no reason to "hide extension types" by default. There's a lot of things I hate about Microsoft, but most of them are difficult to explain. This, however, seems to be something that everyone I have met hates also.

To fix it, just open a file browser window, go to the tools menu, choose preferences.... or something like that. (I can't tell you for sure right now, I'm using linux.)

[QUOTE=Graff;147282]Force manual communication. Reports "unsupported assignment work type".[/QUOTE]I think that the way that P-1's are named is different in v25 vs v24.

IANAP
YMMV
WAG

[quote=Uncwilly;147285]I think that the way that P-1's are named is different in v25 vs v24.

IANAP
YMMV
WAG[/quote]
I Am Not A...Prime? Maybe I Am Not A Prime95 expert or I Am Not A Primeologist or I Am Not A Physicist
Your Mileage May Vary
Wild A** Guess?
Anyway, yeah, I think Uncwilly's right...not so sure what the correct format is.

V5 API typo?
 

[QUOTE=Uncwilly;147285]I think that the way that P-1's are named is different in v25 vs v24.

IANAP
YMMV
WAG[/QUOTE]

IAAP and I was just taking a look at the V5 API and I see an apparent
inconsistency in the documentation. I've included the relevant
portions of the API below:

[CODE]5.4 ra - Register Assignment
...
5.4.2.1 Request Parameters
All common parameters are required and are case-sensitive. Null values are empty strings. Do not use quoted strings.
...
* c - cpu number registering the assignment (integer; 0 is first CPU)
...
7.0 Assignment Work Types

Integer work type values used for ga (Get Assignment), ra (Register Assignment), ap (Assignment Progress), and ar (Assignment Results) API calls.
...
7.2 GIMPS Work Types
...
3 P-1 factoring A=1
b=2
n=<exponent> (integer)
c=-1
B1=<lower_bound> (double)
B2=<upper_bound> (double)
kf=<CSV list of known factors>[/CODE]

Note the use of "c" in work type 3 conflicts with the use "c" in
section 5.4.2.1. The type 3 definition should presumably be
C=-1. (The use of "c" in the other work types should also presumably
be "C".)

This change certainly agrees with the entry in the prime.log
file for the P-1 factoring job I tried to reserve:

[CODE]&c=1&w=3&A=1&b=2&n=...&C=-1&B1=...[/CODE]

Of course, this doesn't help explain the failure to register the
P-1 job. Unless the cgi script adheres strictly to the on-line
API for P-1 jobs... :smile:

The line in my worktodo.add file looked like:

Pminus1=1,2,<exponent>,-1,<B1>,<B2>

Is this not the form that should be used? If not, what
should the form be? TIA for any pointers!

[quote=Uncwilly;147285]I think that the way that P-1's are named is different in v25 vs v24.

IANAP
YMMV
WAG[/quote]
I thought both Pfactor= and Pminus1= are supported in both v25 and v25? (i.e., the difference between letting Prime95 pick B1/B2, or manually specifying them, respectively.)

[QUOTE=Mini-Geek;147287]I Am Not A...Prime? Maybe I Am Not A Prime95 expert or I Am Not A Primeologist or I Am Not A Physicist
[/QUOTE]

I assume that IANAP is "I am not a programmer". At least, that is what
I assumed in my response.

[quote=Graff;147288]The line in my worktodo.add file looked like:

Pminus1=1,2,<exponent>,-1,<B1>,<B2>

Is this not the form that should be used? If not, what
should the form be? TIA for any pointers![/quote]
Ah, I think that's your problem. P-1 assignments can be specified a number of different ways. First of all, the option to specify by Mersenne exponent or by k*b^n+c. For a Mersenne exponent, you'd do it like this:

Pfactor=exponent,factor_depth,is_doublecheck
or, if you'd rather specify B1/B2 yourself:
Pminus1=exponent,B1,B2

For k*b^n+c, you'd do it like this (as you're probably already aware of, just restating this for completeness):

Pfactor=k,b,n,c,factor_depth,number_tests_saved
or
Pminus1=k,b,n,c,B1,B2

However, PrimeNet doesn't assign k*b^n+c P-1 assignments--it only assigns P-1 for Mersenne numbers. Hence, I would expect that unless you use the Mersenne format P-1 line, PrimeNet won't recognize it as a Mersenne number, and thus won't see it as a valid assignment.

Anyway, just a hunch. :smile: