Command-line Update

[hansen_m]

Are there any command-line update options available? Or a standalone installer for the latest version (2.1.4)?

In a large lab environment updating software on systems manually just isn't a viable option.

Thanks,

Matt

[karenacollins]

Hi Matt,
    Unfortunately there is not currently a command line update option. AIJ installation can be achieved by simply copying files to a specific directory structure starting with the top level ImageJ directory. Would it be practical to do an update on one machine and then run a process to copy the resulting ImageJ directory and its sub-directories to all other machines? This is what we do at UofL.
    If that approach is not practical for your environment, I could consider releasing a set of standalone installers version 2.1.5. I am doing final testing on that version now and should be releasing it within the next week.

Karen

On 3/6/2014 10:13 PM, hansen_m [via AstroImageJ] wrote:

Are there any command-line update options available? Or a standalone installer for the latest version (2.1.4)?

In a large lab environment updating software on systems manually just isn't a viable option.

Thanks,

Matt

---

If you reply to this email, your message will be added to the discussion below:

http://astroimagej.1065399.n5.nabble.com/Command-line-Update-tp132.html

To start a new topic under AstroImageJ, email [hidden email]
To unsubscribe from AstroImageJ, click here.
NAML

[karenacollins]

Hi Matt,
 
More specifically, the only 3 files that are updated when using the update feature are:

ij.jar
/plugins/Astronomy_.jar
/macros/StartupMacros.txt

If you run update on one machine and then copy those 3 files to all other machines, you will have the same result as running update on all machines.

Karen

---

If you reply to this email, your message will be added to the discussion below:

http://astroimagej.1065399.n5.nabble.com/Command-line-Update-tp132.html

To start a new topic under AstroImageJ, email [hidden email]
To unsubscribe from AstroImageJ, click here.
NAML

[Matt]

Karen,

Thanks for the quick response!

Yes, we are able to deploy the updated files and we were able to get the update deployed before the morning class started.

I think a standalone installer for each new version or a silent update option would still be really useful.

- Matt

[hansen_m]

Karen,

Do you have a standalone installer for v3.1.0? I'm running into the same issue where I can't easily deploy the latest "self update" version. Another issue is that when using the internal update mechanism the Java app version no longer matches the Mac OS X application bundle version. This makes it difficult for us to report on which version is actually deployed in our labs.

Thanks,
- Matt

[karenacollins]

Hi Matt,

The latest "fresh install" package is for v3.0.0.

I'm guessing you have looked at the release notes (copied below) and determined that some of the v3.1.0 changes are useful in your labs. Assuming that is the case, I will build a Mac version of 3.1.0 tomorrow evening. One difficulty is that I do not currently have direct control of the UofL website page that hosts the AIJ downloads, but hopefully I can get it posted in a day or two. The other issue is that it is a lot of work to build and test all of the individual platform install packages.

Since the update process only installs the three key files ij.jar, Astronomy_.jar, and startup_macros.txt, I'm not sure if the Mac OS X application bundle version issue can be solved for the "update" feature, short of offering an option to disable the update feature. Let me look into that more tomorrow evening as well.

Karen

-------------------------------------------

Version 3.1.0 (with ImageJ 1.47i)

Changes

·         Multi_Plot(MP)

·         Fixed a problem with the “Save all” menu item that caused the aperture overlays shown in the field image saved as a JPG or PNG to be displayed in the wrong location on the underlying image

·         Added a new menu item that creates Minor Planet Center (MPC) data submission format to Multi-plot_Main->File menu (thanks to Ferran Casarramona for the example macro from which the feature was created).

·         Optionally handles the entry of reference star apparent magnitude in the reference star panel. New table columns are created or overwritten when changes to ref star apparent magnitudes are changed. The new columns are Source_AMag_Xxx and Source_AMag_Err_Xxx. The error values are calculated based on the target and ref star photometric error and do not include errors in the ref star magnitudes that are entered. The Source_AMag_Xxx columns are ideal for submitting to the Minor Planet Center.

·         Image Display

·         Added an option to show nearby SIMBAD objects in a web browser when right-clicking within an image with WCS headers

·         Added RGB Composer to Color menu

·         Corrected problem with display of RGB values in the ‘Value’ field above an image

·         Multi-Aperture (MA)

·         Added option to handle apparent magnitude calculations for target stars when magnitudes of reference stars are entered by user

·         Added option to open SIMBAD to objects near the clicked location of a newly defined aperture (when entering apparent magnitude)

·         Added capability to move apertures by clicking and dragging them to a new location

·         Single step now supports two modes of operation after defining apertures. (Option 1) a single click on T1/C1 in each image applies all apertures, performs photometry, and advances to the next image. (Option 2) allows changes to apertures between images but requires a left-click on the T1/C1 star to place apertures. Then after making any aperture corrections,  a right click (or Enter) performs photometry and advances to the next image.

·         Photometer

·         Increased max number of loops on background annulus star removal from 10 to 25 and reduced break threshold from 0.1 to 0.0001.

On 4/24/2015 11:07 AM, hansen_m [via AstroImageJ] wrote:

Karen,

Do you have a standalone installer for v3.1.0? I'm running into the same issue where I can't easily deploy the latest "self update" version. Another issue is that when using the internal update mechanism the Java app version no longer matches the Mac OS X application bundle version. This makes it difficult for us to report on which version is actually deployed in our labs.

Thanks,
- Matt

---

If you reply to this email, your message will be added to the discussion below:

http://astroimagej.1065399.n5.nabble.com/Command-line-Update-tp132p263.html

To start a new topic under AstroImageJ, email [hidden email]
To unsubscribe from AstroImageJ, click here.
NAML

[hansen_m]

Hi Karen,

Yes, a faculty member that uses the software has requested we update to v3.1 because of features he needs for his class. I'm not very familiar with the software beyond just deploying and updating it so I don't know exactly which feature he needs. The option to disable the update feature would be useful because users in our labs wouldn't have permissions to upgrade the software anyway, and that might avoid some confusion but I think we would still get requests from faculty to update to the latest version.

I'm not sure exactly how your self update mechanism works but I think it would make sense to bump the version values of CFBundleVersion and CFBundleShortVersionString listed in the Info.plist file at the end of a successful self update. Having accurate version numbers on applications is how we are able to report on what is currently installed and know when an update is needed. Otherwise we wouldn't be able to differentiate between v3.0 and v3.1 of AstroImageJ. Could the process of building and testing the new individual platform install packages be automated in any way? Shell scripts, CI systems, command line tools, etc.?

I think the crux of it is we deploy hundreds of software packages in our labs and can't always keep up with watching for and deploying all the latest versions. So what we currently do is setup jobs to scrape each web page nightly and look for new versions of software, once found it will automatically download and be prepared in our software distribution system where we can test it and deploy it to the labs soon after. So having those standalone installers for each version helps us be more proactive rather than waiting for a frustrated user to report that a certain piece of software is out of date and scrambling to get it deployed before the class starts.

Thanks for your support. I understand working on this isn't your full time role but I just wanted to give you some background into how it is being used at Penn State and how we deploy it to our labs. It seems to be a very valuable teaching tool for our instructors.

Thanks,

Matt Hansen
Systems Administrator
Penn State University

[karenacollins]

Hi Matt,

I got sidetracked looking for a way to update the Info.plist file when the update feature is used. I think I can see a path now. I would like to get that working before I issue a new fresh install package (hopefully on Tuesday evening).

Karen

On 4/27/2015 1:22 PM, hansen_m [via AstroImageJ] wrote:

Hi Karen,

Yes, a faculty member that uses the software has requested we update to v3.1 because of features he needs for his class. I'm not very familiar with the software beyond just deploying and updating it so I don't know exactly which feature he needs. The option to disable the update feature would be useful because users in our labs wouldn't have permissions to upgrade the software anyway, and that might avoid some confusion but I think we would still get requests from faculty to update to the latest version.

I'm not sure exactly how your self update mechanism works but I think it would make sense to bump the version values of CFBundleVersion and CFBundleShortVersionString listed in the Info.plist file at the end of a successful self update. Having accurate version numbers on applications is how we are able to report on what is currently installed and know when an update is needed. Otherwise we wouldn't be able to differentiate between v3.0 and v3.1 of AstroImageJ. Could the process of building and testing the new individual platform install packages be automated in any way? Shell scripts, CI systems, command line tools, etc.?

I think the crux of it is we deploy hundreds of software packages in our labs and can't always keep up with watching for and deploying all the latest versions. So what we currently do is setup jobs to scrape each web page nightly and look for new versions of software, once found it will automatically download and be prepared in our software distribution system where we can test it and deploy it to the labs soon after. So having those standalone installers for each version helps us be more proactive rather than waiting for a frustrated user to report that a certain piece of software is out of date and scrambling to get it deployed before the class starts.

Thanks for your support. I understand working on this isn't your full time role but I just wanted to give you some background into how it is being used at Penn State and how we deploy it to our labs. It seems to be a very valuable teaching tool for our instructors.

Thanks,

Matt Hansen
Systems Administrator
Penn State University

---

If you reply to this email, your message will be added to the discussion below:

http://astroimagej.1065399.n5.nabble.com/Command-line-Update-tp132p265.html

To start a new topic under AstroImageJ, email [hidden email]
To unsubscribe from AstroImageJ, click here.
NAML

[hansen_m]

Hi Karen,

Once again we have to deploy AstroImageJ to our computer labs this semester. Is it possible to get a standalone build of v3.1.1? Has there been any progress on the idea of providing more frequent builds and issue with updating the Info.plist to reflect the actual version?

[karenacollins]

I made progress on this last summer, but got derailed with other priorities. How soon do you need to install AIJ on your machines? That timing will influence the best way to handle this for you. If I have two weeks, I may be able to get you a fix for the info.plist updates in build of 3.1.1 (or maybe 3.2.0 depending on what else is in the build). If you need it sooner, I'll package up a 3.1.1 OS X version that contains the most recent daily build updates and post that to the installation website.

Karen

On 1/4/2016 12:27 PM, hansen_m [via AstroImageJ] wrote:

Hi Karen,

Once again we have to deploy AstroImageJ to our computer labs this semester. Is it possible to get a standalone build of v3.1.1? Has there been any progress on the idea of providing more frequent builds and issue with updating the Info.plist to reflect the actual version?

---

If you reply to this email, your message will be added to the discussion below:

http://astroimagej.1065399.n5.nabble.com/Command-line-Update-tp132p306.html

To start a new topic under AstroImageJ, email [hidden email]
To unsubscribe from AstroImageJ, click here.
NAML

[hansen_m]

Hi Karen,

Thanks for the quick response. We will need to deploy AstroImageJ in early February so I believe there is time. Fixing the Info.plist version will really help us with deployment as we would then be able to accurately determine which version is installed.

- Matt

[karenacollins]

Hi Matt,

I have released AIJ version 3.2.0 that now updates the Info.plist
entries for Mac installations. The first AIJ "Update" will not change
the values since the old Updater code is performing the download (I
can't do anything to fix that). However, if you "Update" to the same
package twice, the Info.plist values will then be updated. Any future
"Update" operations will update the Info.plist values the first time.

Could you please verify that the double "Update" operation fixes the
Info.plist values as needed on your end?

I have also built new fresh install packages for version 3.2.0. Those
packages will install with the correct Info.plist version values, and
then future AIJ "Updates" will also update the Info.plist values the
first time. Unfortunately, I do not have direct access to post the fresh
install packages, so it may be Monday or Tuesday before the version
3.2.0 files are available for download.

Let me know if this solves your problem.

Karen

[karenacollins]

Hi Matt, The version 3.2.0 installation packages are now available for download here:

http://www.astro.louisville.edu/software/astroimagej/installation_packages/

It would still be helpful to know that the double "Update" process works on your end, if you could try that on at least one machine before installing the new fresh install package.

Thanks!

Karen

[hansen_m]

Thanks, Karen! I just tested this in our labs and everything seems to be working great!