Conky weather+moon scripts (Accuw/WUndergr/NWS/MoonGiant)

TeoBigusGeekus · 2019-10-02 15:18:25

Screw it man, I'll repush the whole thing

Bruno · 2019-10-02 15:52:14

I know... when I pasted the "No-break narrow space" into the message window, it has been erased...
You may have a look to the page https://www.accuweather.com/fr/br/rio-d … ther/45449 and look for "mbar". The character just before mbar is not a space. To check that, just copy/past it into the function:

printf %x "'PAST_HERE"

pressure-unit-separator

I also uploaded my curr_cond file here.

TeoBigusGeekus wrote:

Bruno wrote:
I used printf %x "'"
Normal space:
bruno@Neptune:~$ printf %x "' "
20
Narrow space:
bruno@Neptune:~$ printf %x "'"
202f
I can't replicate this
https://i.imgur.com/MA3QneKt.jpg
Anyway, how did you find out you have this character in your curr_cond file?
I don't seem to have it; I open the file with vs code and vim as well, but still nothing.

Last edited by Bruno (2019-10-02 16:12:32)

loutch · 2019-10-02 16:10:07

i haved the same problem that i use Liberation Sans font

morfik · 2019-10-02 16:38:45

I think I'm orange-addicted guy.

For those who want some decent wallpapers, see here:
https://unsplash.com/collections/179333 … ackgrounds

TeoBigusGeekus · 2019-10-02 18:00:07

@Bruno
I tried with printf and you're right, it is indeed a unicode character.
However, it doesn't show in my conkyrc when I use your curr_cond file

I have the following line in my conkyrc file

override_utf8_locale = true,

I don't know if it makes a difference...

TeoBigusGeekus · 2019-10-02 18:20:54

morfik wrote:

TeoBigusGeekus wrote:
But how am I supposed to apply changes to the master script and then apply them to the branch as well?
You apply changes to the master branch as usual, then switch to the custom_script branch, and cherry pick the commits you made to master branch. If there are any conflicts, it will prompt you to solve them, if not, it will merge the commits. Don't merge the whole branches because this always means the branches will have the same code and can share any new changes you made. Like in the kernel examples, if they merged kernel 5.3 and 5.2 branches, we would have only one kernel code, and not the two, which is not what we want.

Do this test for me, please:
Create a folder with 2 empty files in it, let's say 1 and 2.
Run

git init
git add .
git commit -m "whatever"
git branch test_branch
git checkout test_branch
touch 3
git add .
git commit -m "branch whatever"

You should now have 3 files in your test folder, named 1, 2 and 3.
Then give

git checkout master

Open file #1 and write "Message" in it.
Do

git add .
git commit -m "another whatever"

You now have 2 files (still being in the master branch), with the first one having the string "Message" on its first line.

git checkout test_branch

3 empty files in your folder.

git merge master
git add .
git commit -m "another branch whatever"

3 files with the first having the string "Message" on its first line.

git checkout master

2 files with the first having the string "Message" on its first line.

Why doesn't my script folder behave like this?
Why are the files from one branch visible to the other branch?

morfik wrote:

I think I'm orange-addicted guy.

Go here and choose orange as the color.

bansurveillancecapitalism · 2019-10-02 19:21:24

@teobigusgeekus

Many thanks for the nesting response, but unfortunately it didn't work. I also tried:

{image $HOME/2_WUnderground/WU_Images/$((sed -n '21p' $HOME/2_WUnderground/hourly).png) -p 10,240 -s 50x50}

...but still no joy. I'm sure it is something simple.

FYI

New accuweather wget pull works if connected directly, but does NOT work with my VPN (tried multiple servers with firewall disabled):

Resolving www.accuweather.com (www.accuweather.com)... 2.20.211.239
Connecting to www.accuweather.com (www.accuweather.com)|2.20.211.239|:443... connected.
HTTP request sent, awaiting response...

weather.com: error

convert-im6.q16: unrecognized color `var(--color-icy)' @ warning/color/QueryMagickColorCompliance/2674.
convert-im6.q16: unable to open image `var(--color-icy)': No such file or directory @ error/blob.c/OpenBlob/2701.
convert-im6.q16: no decode delegate for this image format `' @ error/constitute.c/ReadImage/504.
convert-im6.q16: non-conforming drawing primitive definition `fill @ error/draw.c/DrawImage/3265.

All Forecast Images are inaccurate/distorted. I see that the s/var(--color-icy) is NOT in the # Function: colorize svg # section of the script.

Cheers

morfik · 2019-10-02 20:19:08

TeoBigusGeekus wrote:

Do this test for me, please:
...
git merge master

To this point everything is as expected.

Now:

< git:(test_branch 57b1726) > $ git merge master
Merge made by the 'recursive' strategy.
 1 | 1 +
 1 file changed, 1 insertion(+)

< git:(test_branch 60851b2) > $ git status
On branch test_branch
nothing to commit, working tree clean

< git:(test_branch 60851b2) > $ ls -al
total 28
drwxr-xr-x  3 morfik morfik  4096 2019-10-02 21:49:34 ./
drwxr-xr-t 36 morfik morfik 16384 2019-10-01 09:57:20 ../
drwxr-xr-x  8 morfik morfik  4096 2019-10-02 21:49:42 .git/
-rw-r--r--  1 morfik morfik     8 2019-10-02 21:49:34 1
-rw-r--r--  1 morfik morfik     0 2019-10-02 21:47:51 2
-rw-r--r--  1 morfik morfik     0 2019-10-02 21:49:23 3

When you issue `git merge master` , the other two commands don't make any sense. So why do you want to issue the following after the merge?

< git:(test_branch 60851b2) > $  git add .

< git:(test_branch 60851b2) > $ git status
On branch test_branch
nothing to commit, working tree clean

< git:(test_branch 60851b2) > $ git commit -m "another branch whatever"
On branch test_branch
nothing to commit, working tree clean

< git:(test_branch 60851b2) > $ git status
On branch test_branch
nothing to commit, working tree clean

< git:(test_branch 60851b2) > $ ls -al
total 28
drwxr-xr-x  3 morfik morfik  4096 2019-10-02 21:49:34 ./
drwxr-xr-t 36 morfik morfik 16384 2019-10-01 09:57:20 ../
drwxr-xr-x  8 morfik morfik  4096 2019-10-02 21:55:54 .git/
-rw-r--r--  1 morfik morfik     8 2019-10-02 21:49:34 1
-rw-r--r--  1 morfik morfik     0 2019-10-02 21:47:51 2
-rw-r--r--  1 morfik morfik     0 2019-10-02 21:49:23 3

As you can see, it's the same. When I switch to the master branch:

< git:(test_branch 60851b2) > $ git checkout master
Switched to branch 'master'

< git:(master 7aa5e48) > $ git status
On branch master
nothing to commit, working tree clean

< git:(master 7aa5e48) > $ ls -al
total 28
drwxr-xr-x  3 morfik morfik  4096 2019-10-02 21:58:00 ./
drwxr-xr-t 36 morfik morfik 16384 2019-10-01 09:57:20 ../
drwxr-xr-x  8 morfik morfik  4096 2019-10-02 21:58:03 .git/
-rw-r--r--  1 morfik morfik     8 2019-10-02 21:49:34 1
-rw-r--r--  1 morfik morfik     0 2019-10-02 21:47:51 2

And there's two files like before the merge.

TeoBigusGeekus wrote:

Why doesn't my script folder behave like this?
Why are the files from one branch visible to the other branch?

It does. Just create another file to see it for yourself:

< git:(master 7aa5e48) > $ touch 4

< git:(master 7aa5e48✈) > $ git status
On branch master
Untracked files:
  (use "git add <file>..." to include in what will be committed)
        4

nothing added to commit but untracked files present (use "git add" to track)

< git:(master 7aa5e48✈) > $ git checkout test_branch
Switched to branch 'test_branch'

< git:(test_branch 60851b2✈) > $ git status
On branch test_branch
Untracked files:
  (use "git add <file>..." to include in what will be committed)
        4

nothing added to commit but untracked files present (use "git add" to track)

And as you can see `4` showed in both branches. See what git says about this:

< git:(test_branch 60851b2✈) > $ git log --all --oneline --graph --decorate --abbrev-commit


*   60851b2 (HEAD -> test_branch) Merge branch 'master' into test_branch
|\
| * 7aa5e48 (master) another whatever
* | 57b1726 branch whatever
|/
* 9cb454b whatever

Do you see the graph? At the beginning you had one starting point. Then it split into two branches. And in this point you commit changes separately for each of the branch. When you issue `git merge`, all the changes will be merged, and you have again a new starting point for both of the branches (usually after the merge, the other branch can be removed, since it isn't needed anymore). Now you can commit the `4` in the master branch:

< git:(test_branch 60851b2✈) > $ git checkout master
Switched to branch 'master'

< git:(master 7aa5e48✈) > $ git status
On branch master
Untracked files:
  (use "git add <file>..." to include in what will be committed)
        4

nothing added to commit but untracked files present (use "git add" to track)

< git:(master 7aa5e48✈) > $ git add .

< git:(master 7aa5e48+) > $ git commit -m "another another whatever"
[master 244bcd7] another another whatever
 1 file changed, 0 insertions(+), 0 deletions(-)
 create mode 100644 4

< git:(master 244bcd7) > $ git status
On branch master
nothing to commit, working tree clean

< git:(master 244bcd7) > $ ls -al
total 28
drwxr-xr-x  3 morfik morfik  4096 2019-10-02 22:09:34 ./
drwxr-xr-t 36 morfik morfik 16384 2019-10-01 09:57:20 ../
drwxr-xr-x  8 morfik morfik  4096 2019-10-02 22:10:01 .git/
-rw-r--r--  1 morfik morfik     8 2019-10-02 21:49:34 1
-rw-r--r--  1 morfik morfik     0 2019-10-02 21:47:51 2
-rw-r--r--  1 morfik morfik     0 2019-10-02 22:01:04 4

So we have 1, 2 and 4 in the master branch.

Now see what's in the other branch:

< git:(master 244bcd7) > $ git checkout test_branch
Switched to branch 'test_branch'

< git:(test_branch 60851b2) > $ git status
On branch test_branch
nothing to commit, working tree clean

< git:(test_branch 60851b2) > $ ls -al
total 28
drwxr-xr-x  3 morfik morfik  4096 2019-10-02 22:10:11 ./
drwxr-xr-t 36 morfik morfik 16384 2019-10-01 09:57:20 ../
drwxr-xr-x  8 morfik morfik  4096 2019-10-02 22:10:22 .git/
-rw-r--r--  1 morfik morfik     8 2019-10-02 21:49:34 1
-rw-r--r--  1 morfik morfik     0 2019-10-02 21:47:51 2
-rw-r--r--  1 morfik morfik     0 2019-10-02 22:10:11 3

So the branch wasn't touched in any way. Now see the graph:

< git:(test_branch 60851b2) > $ git log --all --oneline --graph --decorate --abbrev-commit

* 244bcd7 (master) another another whatever
| *   60851b2 (HEAD -> test_branch) Merge branch 'master' into test_branch
| |\
| |/
|/|
* | 7aa5e48 another whatever
| * 57b1726 branch whatever
|/
* 9cb454b whatever

Do you see? They split again. Does this make any more sense to you now?

Last edited by morfik (2019-10-02 20:22:55)

Bruno · 2019-10-02 20:50:42

I also have override_utf8_locale= true...
I tried removing "font= 'Ubuntu:style=medium:size=9'"

before:
unicode

after:
unicode

The no-break narrow space character seems to be missing in Ubuntu font.

TeoBigusGeekus wrote:

@Bruno
I tried with printf and you're right, it is indeed a unicode character.
However, it doesn't show in my conkyrc when I use your curr_cond file
https://i.imgur.com/cjgRfdQt.jpg
I have the following line in my conkyrc file
override_utf8_locale = true,
I don't know if it makes a difference...

TeoBigusGeekus · 2019-10-02 21:32:35

Yes, I totally understand this and that's the way I expect it to behave.
Also, that's what I like 1_Accuweather script to do:
I have the master branch, the one you guys download in order to use, and the random_photo branch, in which I've added a few lines in the script to show a random pic from my pictures folder in my conky.
I want to make changes to the master branch, ie. corrections to the script, and then merge it to the random_photo one.
That's what I used to do until now and it worked exactly as I wanted it to work, ie:
MASTER BRANCH
code, code, code,
.......
.......
code, code, code

RANDOM PHOTO BRANCH
code, code, code,
.......
My custom script lines
.......
code, code, code

Oops, Accuweather's screwed up again. Switch to the master branch and correct the script
MASTER BRANCH
corrected code, corrected code, corrected code,
.......
.......
corrected code, corrected code, corrected code

then switch to random_photo, do a git merge master and end up with
RANDOM PHOTO BRANCH
corrected code, corrected code, corrected code,
.......
My custom script lines
.......
corrected code, corrected code, corrected code

That's exactly how I want it to behave and that's exactly how it behaved until now.
Whenever I switched from branch to master, the downloaded curr_cond_raw, curr_cond, hourly_raw, etc. files disappeared, as they should. I worked on the master script, committed the changes, switched to the branch (the curr_cond, etc. files reappeared) and merged the master to it.
Both my master script and the one I use on my pc are corrected.

But from this morning on, if I switch to the master branch, delete all forecast images, daily_forecast, raw files, etc. and then switch to the branch, the files remain deleted.
Also, if I run the script from the branch, as I already do, the downloaded files remain there if I switch again to the master branch.
That's what's changed from yesterday, every file I create or remove from one branch affects the other branch as well.
I want it to be as it was before, ie. if I create a file called altogether24_2016 in the random_photo branch, I don't it to be there when I switch to the master branch.
Sorry, you might be witnessing a personal brain fart, but I really am confused.

Last edited by TeoBigusGeekus (2019-10-02 21:47:47)

TeoBigusGeekus · 2019-10-02 21:34:46

bansurveillancecapitalism wrote:

@teobigusgeekus
Many thanks for the nesting response, but unfortunately it didn't work. I also tried:
{image $HOME/2_WUnderground/WU_Images/$((sed -n '21p' $HOME/2_WUnderground/hourly).png) -p 10,240 -s 50x50}
...but still no joy. I'm sure it is something simple.

FYI
New accuweather wget pull works if connected directly, but does NOT work with my VPN (tried multiple servers with firewall disabled):
Resolving www.accuweather.com (www.accuweather.com)... 2.20.211.239
Connecting to www.accuweather.com (www.accuweather.com)|2.20.211.239|:443... connected.
HTTP request sent, awaiting response...

weather.com: error
convert-im6.q16: unrecognized color `var(--color-icy)' @ warning/color/QueryMagickColorCompliance/2674.
convert-im6.q16: unable to open image `var(--color-icy)': No such file or directory @ error/blob.c/OpenBlob/2701.
convert-im6.q16: no decode delegate for this image format `' @ error/constitute.c/ReadImage/504.
convert-im6.q16: non-conforming drawing primitive definition `fill @ error/draw.c/DrawImage/3265.
All Forecast Images are inaccurate/distorted. I see that the s/var(--color-icy) is NOT in the # Function: colorize svg # section of the script.
Cheers

Which VPN do you use? It works with my expressvpn.

What distro do you use? It seems that the weather.com script only works on Arch...

@Bruno
Yep, seems like a change of font is a good idea.

Last edited by TeoBigusGeekus (2019-10-02 21:36:12)

TeoBigusGeekus · 2019-10-02 21:42:16

bansurveillancecapitalism wrote:

@teobigusgeekus
Many thanks for the nesting response, but unfortunately it didn't work. I also tried:
{image $HOME/2_WUnderground/WU_Images/$((sed -n '21p' $HOME/2_WUnderground/hourly).png) -p 10,240 -s 50x50}
...but still no joy. I'm sure it is something simple.

Wait a minute. Why do you wanna do this?
The script has already done this for you; all the next hours' images are in the Forecast_Images folder.
The first image, ie. the one reported on the 21st line of the hourly file, is the h1.png file.
So, all you have to do is

{image $HOME/2_WUnderground/WU_Images/Forecast_Images/h1.png) -p 10,240 -s 50x50}

TeoBigusGeekus · 2019-10-02 22:02:35

@morfik
Ok, I have some progress to report.
It only happens with the script output files: curr_cond, daily_forecast, etc. and not with randomly generated files.

morfik · 2019-10-03 00:20:40

Does that happen without any changes to any of the branches? So it worked before, now you wake up, and stopped working? That's really weird, and I don't really know what might cause this. I've never seen something like this before. Maybe you changed something, like .gitignore?

TeoBigusGeekus · 2019-10-03 04:48:45

morfik wrote:

Does that happen without any changes to any of the branches? So it worked before, now you wake up, and stopped working? That's really weird, and I don't really know what might cause this. I've never seen something like this before. Maybe you changed something, like .gitignore?

It all started after I did that git push -u origin master without having first switched to the master branch.
So I pushed the master script into the master branch from the custom_script branch.
Afterwards, the

curr_cond
curr_cond_raw
daily_forecast
hourly
hourly_raw1
hourly_raw2
altogether24
altogether24_2015
altogether24_2016

stay the same when I switch between branches.
I thought the custom_script branch was to blame, so I deleted it and created a new one called random_photo.
Nope, still the same behavior.
The good thing is that if I create a random file in either of the branches, add it and them commit the changes, it doesn't show up when I switch branches.
The bad thing is that I want this thing to happen again for the aforementioned files as well.

PS: Gitignore is untouched for both branches.
PS2: I did a

git ls-tree --full-tree --name-only -r HEAD

to both branches and the previous file list is not tracked by git. All the files are into the .gitignore file.

Last edited by TeoBigusGeekus (2019-10-03 05:02:26)

TeoBigusGeekus · 2019-10-03 05:08:57

Bruno wrote:

I also have override_utf8_locale= true...
I tried removing "font= 'Ubuntu:style=medium:size=9'"
before:
https://i.ibb.co/d23Bdh5/pressure-unit-separator.png
after:
https://i.ibb.co/qRWM1Q2/pressure-unit-separator-ok.png
The no-break narrow space character seems to be missing in Ubuntu font.

I found this.

bansurveillancecapitalism · 2019-10-03 05:54:06

Thanks again for the response.

Wait a minute. Why do you wanna do this?
The script has already done this for you; all the next hours' images are in the Forecast_Images folder.
The first image, ie. the one reported on the 21st line of the hourly file, is the h1.png file.
So, all you have to do is
{image $HOME/2_WUnderground/WU_Images/Forecast_Images/h1.png) -p 10,240 -s 50x50}

As I mentioned in a previous post: For some reason, the "10days" file in WUnderground folder does NOT populate, but the "curr_cond" and "hourly" do populate, as expected. Perhaps it is the pull command, no idea. The script does generate a 10days file, but it is empty. So, as a workaround, I wanted to specify the variable in a day, or days, ahead that resides in a specific line in the populated hourly file to grab the icon image specified after the "icon_extd" line.

Have not tested the issue without disabling the VPN or firewall. Will give it a try sometime today.

Cheers

TeoBigusGeekus · 2019-10-03 05:59:12

Oh OK, I missed that, sorry.
Try running the script (just the script) from the command line and see if you get any error messages.

ohnonot · 2019-10-03 07:21:54

TeoBigusGeekus wrote:

Shit guys, has Accuweather done it again?
Is w3m banned now? (along with wget, curl and lynx?)
EDIT: And apparently elinks as well

You always should test with different user agents!
(apparently w3m does not have that option, but lynx has, and wget and curl of course).

I thought I might try to fiigure out how to build it dynamically against the installed browser.

morfik · 2019-10-03 12:53:38

@TeoBigusGeekus:
Could you show the output of:

$ git remote update && git status
$ git log --all --oneline --graph --decorate --abbrev-commit

#2841 2019-10-02 15:18:25

Re: Conky weather+moon scripts (Accuw/WUndergr/NWS/MoonGiant)

#2842 2019-10-02 15:52:14

Re: Conky weather+moon scripts (Accuw/WUndergr/NWS/MoonGiant)

#2843 2019-10-02 16:10:07

Re: Conky weather+moon scripts (Accuw/WUndergr/NWS/MoonGiant)

#2844 2019-10-02 16:38:45

Re: Conky weather+moon scripts (Accuw/WUndergr/NWS/MoonGiant)

#2845 2019-10-02 18:00:07

Re: Conky weather+moon scripts (Accuw/WUndergr/NWS/MoonGiant)

#2846 2019-10-02 18:20:54

Re: Conky weather+moon scripts (Accuw/WUndergr/NWS/MoonGiant)

#2847 2019-10-02 19:21:24

Re: Conky weather+moon scripts (Accuw/WUndergr/NWS/MoonGiant)

#2848 2019-10-02 20:19:08

Re: Conky weather+moon scripts (Accuw/WUndergr/NWS/MoonGiant)

#2849 2019-10-02 20:50:42

Re: Conky weather+moon scripts (Accuw/WUndergr/NWS/MoonGiant)

#2850 2019-10-02 21:32:35

Re: Conky weather+moon scripts (Accuw/WUndergr/NWS/MoonGiant)

#2851 2019-10-02 21:34:46

Re: Conky weather+moon scripts (Accuw/WUndergr/NWS/MoonGiant)

#2852 2019-10-02 21:42:16

Re: Conky weather+moon scripts (Accuw/WUndergr/NWS/MoonGiant)

#2853 2019-10-02 22:02:35

Re: Conky weather+moon scripts (Accuw/WUndergr/NWS/MoonGiant)

#2854 2019-10-03 00:20:40

Re: Conky weather+moon scripts (Accuw/WUndergr/NWS/MoonGiant)

#2855 2019-10-03 04:48:45

Re: Conky weather+moon scripts (Accuw/WUndergr/NWS/MoonGiant)

#2856 2019-10-03 05:08:57

Re: Conky weather+moon scripts (Accuw/WUndergr/NWS/MoonGiant)

#2857 2019-10-03 05:54:06

Re: Conky weather+moon scripts (Accuw/WUndergr/NWS/MoonGiant)

#2858 2019-10-03 05:59:12

Re: Conky weather+moon scripts (Accuw/WUndergr/NWS/MoonGiant)

#2859 2019-10-03 07:21:54

Re: Conky weather+moon scripts (Accuw/WUndergr/NWS/MoonGiant)

#2860 2019-10-03 12:53:38

Re: Conky weather+moon scripts (Accuw/WUndergr/NWS/MoonGiant)

Board footer