Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.eso.org/projects/dfs/team/JP2PP-test-report-V2-2-beta5.txt Дата изменения: Tue Nov 28 20:24:06 2000 Дата индексирования: Sun Apr 13 22:54:41 2008 Кодировка: Поисковые слова: annular solar eclipse

TEST REPORT FOR P2PP V2.2beta5
==============================

The main goal here was to check a new release P2PP implementing
several fixes on the current official V2.1.1 (in particular, the DFS PR00801).


Remarks from KH and answers from AMC
====================================

Remarks 1/ and 2/ were minor questions.


> 3/ HIGH: create a FORS2 OB, export it. Check it IN it. Check it OUT,
> and re-export it (with a specific name).
> Then, compare the 2 OBX files: following difference:
> < ObsGrade ""
> ---
> > ObsGrade " "
> 22c22
> < objectClass " Moon "
> ---
> objectClass " Unknown "
> 29c29
< description "rwerwe "
> ---
> description "rwerwe"
>
> If I View the OB after check-in/check-out, value is indeed re-set to
> "Unknown".

Looks like a bug: item 1106 (for the V2.2 branch) and 1145 (for the V2.1.x branch).

> 4/ HIGH : When running an action (e.g. Synchronise, ...), a clock is
> displayed to indicate that P2PP is working : I noticed that this clock
> is still displayed EVEN if the action is ended, but if I move the mouse,
> the cursor immediatly goes back to its normal form. Just a small problem
> of removing this clock, once the action is performed, but should be
> fixed in the next beta I think, because maybe confusing for the user.

Looks like a swing problem, I don't know if we can do anything
about it. Anyway: item 1107.

> 5/ synchronizing OBs : does the "User Priority" belong to the OD ? If
> so, there is a problem, since it's not correctly synchronized (I found a
> difference between 2 synchronized OBs).

No. User priority does not belong to the OD, but to the OB itself. There
is currently no way to synchronize that, or the user comments, etc.

> 6/ CRITICAL : Well, in fact the Java exception ...
> is ALSO originated when login as a
> different user *during the same P2PP session* : it should not, isn't it
> ? otherwise it's confusing, ...

The Java exception is OK (in fact, it shouldn't need to be printed: it
means "file not found"). However we should change the message here, to
include something like "IF this is the first time you use this account,
THEN you can ignore this warning"; item 1108.

> And then, what is more problematic, is that login again during the same
> session does NOT work : following Java exceptions raised, progress bar
> blocked, and P2PP hanged.

This instead is a bug, item 1109.

> 7/ more about point 6/
> Quitting P2PP, re-starting it under the first username : another Java
> exception:
> java.io.FileNotFoundException: /diska/visitor/cache/lockinfo.1797
> at java.io.FileInputStream.(FileInputStream.java)
> ...
> And the warning message as above : what should I do here to avoid this
> message ?

That could be a bug, unless you exited with Ctrl-C. Can you check?

> 8/ point 6/ again:
> Under my cache directory, there are 2 files named
> unlocked.52005 and unlocked.52001
> Should I remove them ?

No! They are the "unlock" files (we can't create "lock" files). If you
delete them, you'll get the warning message again.

> 9/ more about previous points 6/ to 8/ : in fact it is not enough to
> remove the unlocked.52005 and unlocked.52001 (I get the same warning
> message and java exception in that case). I needed to exit from my UNIX
> session (user Visitor) and re-login again to avoid this
> message/exception.

I hope this is clearer now. If not, please give me a call.

> 10/ MINOR : ITEM 0883 : better erro message displayed when p2pp-db.cf
> not found. But when this config file is here and NOT readable, the error
> message says : "NO WRITES TO READ ../p2pp-db.cf". Better to say "NO
> RIGHT TO READ ..." !

Item 1110.

> 11/ MINOR ? ITEM 0914: illegal values in template grid : OK it turns in
> red if value is incorrect. But, the Right ascencion (in Targte Package
> area) may contain a wrong value 23 (instead of 23:00:00.000).

This is a display problem: the value is stored OK internally, as you can
see if you select another OB, that that one again. Item 1111.

> 12/ CRITICAL : ITEM 0914 again : improperly fixed I think.
> Try to enter a wrong value for GS RA (25 where it should be lower than
> 24, for the FORS2-066_ech_acq_mxu template) : the wrong value is not in
> red, and no way to quit this field.
> And in fact P2PP is frozen .. I can not do anything but just Quit P2PP.
> The field is not editable anymore. Java exception raised several times :
> java.lang.IllegalArgumentException: Illegal time value at
> org.eso.ohs.gui.docview.RaAdaptor.checkBounds(RaAdaptor.java:91) at
> org.eso.ohs.gui.docview.DegreeAdaptor.convertToMilliArcSeconds(DegreeAdaptor.java:128) at
> org.eso.ohs.gui.docview.RaAdaptor.toMetaLevelObject(RaAdaptor.java:50) at
> org.eso.ohs.p2pp.views.EditorFactory$CoordTextField.getValue(EditorFactory.java:1055) at
> org.eso.ohs.p2pp.views.EditorFactory.getCellEditorValue(EditorFactory.java:161)
> at javax.swing.JTable.editingStopped(JTable.java:2556) at
> javax.swing.DefaultCellEditor.fireEditingStopped(DefaultCellEditor.java:281) at
> javax.swing.DefaultCellEditor.stopCellEditing(DefaultCellEditor.java:239) at
> org.eso.ohs.p2pp.views.ODTemplateTable.stopCellEditing(ODTemplateTable.java:278) at
> org.eso.ohs.p2pp.views.ODTemplateTable$TableMouseListener.saveChanges(ODTemplateTable.java:522) at
> org.eso.ohs.p2pp.views.ODTemplateTable$TableMouseListener.mouseExited(ODTemplateTable.java:563) at
> java.awt.AWTEventMulticaster.mouseExited(AWTEventMulticaster.java:247) at
> java.awt.AWTEventMulticaster.mouseExited(AWTEventMulticaster.java:247) at
> java.awt.AWTEventMulticaster.mouseExited(AWTEventMulticaster.java:247) at
> java.awt.AWTEventMulticaster.mouseExited(AWTEventMulticaster.java:247) at
> java.awt.AWTEventMulticaster.mouseExited(AWTEventMulticaster.java:247) at
> java.awt.AWTEventMulticaster.mouseExited(AWTEventMulticaster.java:247) at
> java.awt.Component.processMouseEvent(Component.java:2364) at
> java.awt.Component.processEvent(Component.java:2203) at
> java.awt.Container.processEvent(Container.java:901) at
> java.awt.Window.processEvent(Window.java:367) at
> java.awt.Component.dispatchEventImpl(Component.java:1812) at
> java.awt.Container.dispatchEventToSelf(Container.java:955) at
> java.awt.LightweightDispatcher.retargetMouseEvent(Container.java:1839) at
> java.awt.LightweightDispatcher.trackMouseEnterExit(Container.java:1706) at
> java.awt.LightweightDispatcher.processMouseEvent(Container.java:1600) at
> java.awt.LightweightDispatcher.dispatchEvent(Container.java:1531) at
> java.awt.Container.dispatchEventImpl(Container.java:933) at
> java.awt.Window.dispatchEventImpl(Window.java:509) at
> java.awt.Component.dispatchEvent(Component.java:1744) at
> java.awt.EventDispatchThread.run(EventDispatchThread.java:82)

To be fixed


> 13/ more about point 12/ : if I enter wrong values for GS DEC, RA/DEC
> Velocity, exposure time filed/slit : OK it works : turned in red. And
> the famous "red dot" appears in the main OB grid (was NOT the case
> above). I tried entering wrong values for several templates : OK except
> the GS RA seems to work, but I can not check ALL my templates ... !

To be fixed


> * MORE ABOUT ITEM 1107 :
>
> > Hmm, looks like a swing problem, I don't know if we can do anything
> > about it. Anyway: item 1107.
>
> Well, the more I'm using this V2.2, the more I think this is disturbing,
> because the user has to move the mouse or click somewhere very often to be
> sure that the action is really finished.
> Something should be done here : either there is a way to reset the cursor
> once the action is performed, or display some message, ... Otherwise, it can
> really be disturbing sometimes.

Item 1107.


> * MORE ABOUT ITEM 1108 :
>
> I don't like this solution very much : first of all, adding another message
> to the existing one may really be confusing. What does "the first time" means
> : the first time for the current UNIX session ? the first "absolute" time for
> the user ? the first time since the current P2PP release has been installed ?
> the first time since the "cache" directory has been created for the current
> user ?

The first time you run V.2.2 you'll get that warning, but only the first
time. From then on you only get that if indeed two users are using the
same Unix account and the same P2PP account at the same time.


> And then, I would prefer to improve the login-check algorithm instead.

For which you have to include some OS-specific code (there is no concept
like PID or user ID doe Windows/MacOS). Item 1115.

> Another problem: what happens if P2PP crashes or the machine where it's
> running crashes, or for any reason, the "lockinfo" file is not properly
> deleted. This lockinfo will have then to be removed manually, but I'm not
> sure that it's up to the user to do that and to know things about the cache
> directory and these lock/unlock files ...

In fact one should *never* delete that file: absence of that file means
precisely that a user is active.
If P2PP crashes the user will be displayed that warning again -- there
is no way to avoid that 100% (see Netscape).

> Finally, I have a lot of questions about the security level needed for P2PP :
> maybe it's an issue requiring a specific meeting with other people like Dave
> ... ?

Yes, we have no new requirements for security so far.

> * MORE ABOUT ITEM 0914:
>
> Checked again with Bob yesterday afternoon and there is indeed a problem here
> : wrong value for GS RA freezes somehow P2PP (no way to click somewhere else,
> arrows keys don't work, the same for RETRUN; but TAB works). I checked some
> of other fields (no problem there), but it will be better if you checked also
> on your side : maybe this problem is also concerning other template fields
> ...

Item 1116.

> 14/ previously I only needed to attach an acquisition template to an Obs
> Block to get it as valid (no error detected, no red dot). Now, if I do the
> same it complains : I also need to attach a science template or a calibration
> one, to get a valid OB : is it normal ? If so, should be clearly said in the
> User's Manual.

My fault: this was item 0871 "An OB with an empty OD is invalid and
should be be rejected", requested by Dave. Somehow it didn't make it
into the release notes.

> 15/ checking item 0819 (new IMPEX warning messages if IMPEX file contains
> errors).
> - OK it works when same name is imported (message displayed).
> - if keyword/value pair is repeated, the last one is used, but NO WARNING
> MESSAGE displayed as it should ?
> - if a wrong value format is entered in the obx file, OK, it complains and
> set it with the default value.
> - if an unknown keyword is specified, does NOT complain as it should ?

Item 1117.

> 18/ HIGH : Having a wrong TSF ( FORS2_mxu_mnt_createMio.tsf) where
> TPL.INSTRUM is set to FORSX (instead of FORS2).
> When I run P2PP, TSF are recomputed and I get an error message saying "
> Invalid templates in instrument package(s). The following instrument
> package(s) FORS2 have invalid template signatures. Please use engineering
> mode ...".
> Problem : the list of wrong TSF is NOT displayed here. It is by the debug
> option, but not at the user level.

I think we can reject this comment.
End-users should never modify TSFs: if they do, they're on their own, we
do not support them.
Engineering users are told (on the GUI) that some TSF is invalid, and
they should use Engineering Mode to find out what is wrong: I think this
is enough.

19/ ITEM 1089: OK I found an ISAAC template with a "keywordlist", and indeed if no value is entered,
NODEFAULT is now returned ==> FIXED.

20/ITEM 1087: I understand also: double-click opens a Select from list box containing exactly the value
defined in the ISAAC isf file ==> FIXED

Item 1118.


> 21/ HIGH ? something new about the ALIAS mechanism:
> a) I updated the ISAAC ISF file to add a new value for an ALIAS
> (already pointing to 4 values).
> I created an ISAAC OB with the template "ISAACLW_all_cal_Darks" which
> uses this ALIAS.
> Double-click on the "Detector Readout Mode" field opens a "Select from
> list" box : OK I see the previous 4 values and the new one I just added
> in the ISF file.
> Add this new value to the value list associated with this field : OK it
> works. Duplicating this OB shows no difference.
>
> b) Doing quite the same sequence but creating also a new keyword this
> time by using the "add keyword" field and the "Add keyword" button of
> the "Select from list" box.
> Close this "Select from list" box and the OB View, shows that this ISAAC
> OB was correct : no red dot.
> But, when I duplicated it, the new duplicated OB got a red dot !
> Duplicating again several times : the same, all new duplicated OBs are
> wrong (red dot), the original one is still OK.
> Apply "Verify" on these wrong OBs says "there is one error in the OD",
> and the OB View shows indeed the "Detector Readout Mode" field value in
> red.
> If I double click again on this field, delete the last value (the last
> keyword added via the "Select from list" box), the red dot disappears.
>
> ==> seems like adding a new ALIAS keyword directly updating via the ISF
> file works, but creating/adding new ALIAS keyword from the "Select from
> list" does NOT work ...

Item 1118 (for the V2.2 branch) and 1146 (for the V2.1.x branch).



CONCLUSION
==========
This release contains CRITICAL bugs which need first to be fixed before installation on the mountain.