| ID |
Date |
Icon |
Author |
Author Email |
Category |
OS |
ELOG Version |
Subject |
|
1803
|
Mon Apr 10 20:08:02 2006 |
 | Yoshio Imai | | Question | Linux | 2.6.1-1671 | Re: elog client authentication and attachment comment |
| Stefan Ritt wrote: | | We could save the username/password in a file on the server, which is maybe readable only by the owner. |
I have discussed it with the others, and it sounds like a good idea. There is only the debate whether it should be readable by the owner or by the root user of the elog server. I can't tell at the moment which is more favourable ... |
|
1981
|
Wed Oct 11 16:03:52 2006 |
| Yoshio Imai | | Question | Linux | 2.6.2-1714 | Re: Turn off smileys? |
| Stefan Ritt wrote: | Put a \ in front of anything which is turned into a smiley, so ;) gets converted into  , but \;) does not get converted |
Other suggestion: What about going the other way round and making this \?) the smiley and this ?) the usual question in brackets? If there was a special sequence to announce the unusual case (i.e. the smiley), I think less people would complain about having unwanted conversions ... |
|
2044
|
Mon Nov 6 17:41:13 2006 |
 | Brian Marshall | | Request | Windows | 2.6.2 | XML, CSV and Raw export produces 'last' page only | The Find command page provides an option, 'Display n entries per page' which is useful when the result of the search is to be viewed on-screen.
However this value is also applied to CSV, XML and Raw export and only the 'last' page of data is exported.
For example, if there are 17 matching entries and n is set to 8, only one entry will appear in the exported file.
In my opinion, this option should be ignored for XML, CSV and Raw export and all matching entries should be exported. |
|
2183
|
Tue Apr 3 13:17:27 2007 |
 | Yoshio Imai | | Request | Linux | 2.6.4-1795 | Multiple ideas for multiple logbooks | Hi!
After a long pause, here we are again with a bunch of new ideas (and problems ...;) )
1. Multiple login
Since we are using the elog system now for most aspects of our work, we are more and more often confronted with the need to work with different logbooks open in parallel (e.g. the shift logbook as well as the personal analysis logbook, via tabbed browsing). These logbooks do not always have the same users allowed to access them, so when having logged on to one logbook and then opening another one in a different browser tab, I have to log on as a different user, thereby losing the login on the first logbook. Is there any way to change the cookie structure so as to allow one (physical) user to be logged on as more than one elog user at a time?
2. <Ctrl-T>
While using this logbook (and tabbed browsing), I noticed that the keyboard shortcut <Ctrl-T> now leads to the creation of a table (great thing, btw). However, many of our users are accustomed to this shortcut opening a new browser Tab. Is there any way to disable these elog shortcuts or otherwise disentangle these functions (by using <Alt> instead of <Ctrl> or something similar, since the functionality itself is very good)?
Thanks for any idea on these issues, and thanks for the great software! |
|
2187
|
Tue Apr 3 17:55:21 2007 |
 | Yoshio Imai | | Bug report | Linux | 2.6.4-1826 | Re: Multiple ideas for multiple logbooks |
| Stefan Ritt wrote: | Well, you didn't realize, but you were my beta-tester for the keyboard shortcuts  . I didn't yet publish it, so you must have gotten this brand new feature from SVN. |
You should really have told me . Actually, I found out about the <Ctrl-C> issue while using this logbook.
| Stefan Ritt wrote: | | So get the SVN update, and let me know what you think |
Did so, we are running now -1826. I think it is very usable, with maybe one exception (this might be a browser issue, however): When using the shortcuts and/or the ELCode buttons, sometimes the code is generated and the cursor in between the [XX] and [/XX] directive, but sometimes, the whole code directive is selected as if for editing
which prevents from immediately continuing to type (which, I guess, was the idea behind the shortcuts). The behaviour is not completely reproducible, and I have even seen it change during one and the same browser (=firefox) session while working on the same entry (*). I also had it once that the browser out of a sudden completely ignored the ELCode meaning, and instead used <Ctrl-I> for page info and <Ctrl-B> for the history.
(*) It just happened again!
BTW, there are (of course ) two other issues. They are still present in -1826
- one bug was reported by a logbook user: when using the "find" function in a logbook, and selecting one value from the "Author" drop-down list and another from the "Category" drop-down list, the elogd reports no entries found, but when specifying only the "Author" and leaving "Category" blank, many entries are found, including those with the "Category" being looked for. When specifying only the "Category" and leaving the "Author" blank, the elogd also reports no entries found. Has anyone reported a similar behaviour?
- We have one logbook top group for the beamtimes. Under certain conditions, a preset text is to be displayed.
{2&a} Preset Text = resource/text_1.txt
{2&b} Preset Text = resource/text_2.txt
However, the definitions of these conditions have to be repeated for every logbook in the group (putting it only under [global beamtimes] doesn't do it), otherwise they are ignored.
Any idea on these? |
|
2191
|
Wed Apr 4 11:37:56 2007 |
 | Yoshio Imai | | Request | Linux | 2.6.4-1795 | Re: Multiple ideas for multiple logbooks |
| Stefan Ritt wrote: | | Actually there is a way |
... and it works! Thank you!
| Stefan Ritt wrote: | | I fixed that now in the updated elcode.js. |
That works now, too!
| Stefan Ritt wrote: | | Your other problem of ignoring the control keys at all I could not reproduce. Try to reproduce it in a well-defined way and let me know. |
I will try to do that (new beamtime, new occasions ... ).
| Stefan Ritt wrote: | | Maybe you used a userlist attribute for the author? |
No, I didn't. I will try to investigate this further and let you know if I find anything.
So for the moment ... thanks for the help! |
|
2193
|
Wed Apr 4 14:20:41 2007 |
| Yoshio Imai | | Bug report | Linux | 2.6.4-1826 | Re: Multiple ideas for multiple logbooks |
| Stefan Ritt wrote: | | Maybe some other problem? |
Possible, but I think it worked also on our server until the number of logbooks in the top group got larger than one. Can you try to see if your code still works when there is more than one logbook defined? |
|
65761
|
Thu Feb 28 19:07:19 2008 |
| Yoshio Imai | | Request | Linux | All | 2.6.5 | #include statements and attachment visibility | Hi!
First of all, thank you again for the great software and all the support.
Recently, one collaborator here noted that it would be helpful if the preview of attached files could be disabled on a file-by-file basis (via a checkbutton next to the "Upload" button maybe?). This applies e.g. to cases where someone performs a measurement outside of routine operations and attaches the ASCII data file (preview not wanted, in particular if it contains many lines) and the graph representing the evaluation (preview wanted). The disabling should apply to both single-entry view and list view with "Show attachments" option.
Another "fancy" idea of ours would be to allow #include-like statements in the ELOG config file. E.g. if the number of logbooks gets large, people might choose to put old logbooks to an archive disk which is then stored on some shelf. If a user then wants to access these, the disk could be mounted again (say, under /elog-archive). But since we don't know which archive disk has been mounted, and in order to keep the main config file small, the best would be to have the configurations for the logbooks of each disk on the disk itself (say, in a file called additional.config). We could then have a line like#include /elog-archive/additional.config in the main config file. When the elogd is (re)started, it would try to include that file. If it finds none (because no archive disk is mounted) it would silently ignore this. But if it finds such a file, it would include the logbook definitions found therein.
Do you think it is possible (and preferable) to implement this?
Cheers
Y |
|