Ok.
If it's a permission issue, does it help to run Apertium viewer as root?
Den 18. jan. 2017 1.50 PM skrev "Jacob Nordfalk" <[email protected]>:
> Hi Robert,
>
> Thanks for the bug report. You forgot to send it to the list.
>
> So basically
>
> - the file is there
>
> - you can read the content of the file (try 'cat /home/rob/Deployed/
> apertium/eng_lanc/apertium-eng_lanc/eng_lanc.automorf.bin')
>
> but apertium-viewer cannot read the file.
>
> I have no idea how that could happen, but in that case next step would be
> to see is there is any 'system' in that some programs are not being able to
> read the content of the file, and others are. Can other Java programs
> access the contents of the file?
>
> As I said I dont have time now to find and correct bugs.
> And, its completely OK to tell about your problems on the list - no
> excuses are requred :-)
>
>
> switching to the C++ version did not help, though
>> the fails were different (perhaps the problem is not as simple as file
>> permissions?).
>
>
> Here the error messages/log would be the only way for me to tell.
>
> Yours,
> Jacob
>
>
> 2017-01-17 12:06 GMT+01:00 andre street <[email protected]>:
>
>> Hello Jacob,
>>
>> I still feel I owe you apology for posting unhappy comments to the list.
>> The computer I use for Apertium can not run a development environment
>> for Java. A JDK running a program uses most of the memory. But here is a
>> more detailed bug report. If there is a more appropriate place to send
>> this report, also I apologise (I did not expect to develop with
>> Apertium :) )
>>
>> == ERROR
>> apertium-viewer fails in many situations reporting it is unable to read
>> files.
>>
>> == ENVIRONMENT
>> X64 Linux Mint OpenJDK 7
>> Java version reported as 1.7
>>
>> == CODE SOURCE
>> From:
>> https://svn.code.sf.net/p/apertium/svn/trunk/apertium-viewer
>>
>> Inside this folder is a pre-built jar.
>>
>>
>> == STEPS TO REPLICATE BUG
>> 1. Launch the jar
>> Either from commandline or direct invokation of the Java Runtime
>>
>> Local language files are correctly discovered.
>>
>> 2. Apertium viewer first loads a 'local' pair.
>> Which reports this error,
>>
>> java.io.FileNotFoundException:
>> '/home/rob/Deployed/apertium/eng_lanc/apertium-eng_lanc/eng_lanc.automorf.bin'
>> (No such file or directory)
>> at java.io.RandomAccessFile.open(Native Method)
>> at java.io.RandomAccessFile.<init>(RandomAccessFile.java:241)
>> at org.apertium.utils.IOUtils.memmap(IOUtils.java:495)
>> at org.apertium.utils.IOUtils.openFileAsByteBuffer(IOUtils.java
>> :322)
>> at org.apertium.lttoolbox.LTProc.doMain(LTProc.java:224)
>> at org.apertium.pipeline.Dispatcher.doLTProc(Dispatcher.java:180)
>> at org.apertium.pipeline.Dispatcher.dispatch(Dispatcher.java:264)
>> at apertiumview.Pipeline$PipelineTask.run(Pipeline.java:129)
>> at apertiumview.Pipeline$1.run(Pipeline.java:40)
>>
>> == CROSS-CHECKING
>> '/home/rob/Deployed/apertium/eng_lanc/apertium-eng_lanc/eng_lanc.automorf.bin'
>> exists.
>>
>>
>> == TESTS
>> 1. Change permissions
>>
>> permissions on the file:
>> owner:rw group:r other:r
>>
>> Let's open some permissions,
>> permissions on the file:
>> owner:rw group:r other:rw executable:true
>>
>> Same error:
>>
>> java.io.FileNotFoundException:
>> '/home/rob/Deployed/apertium/eng_lanc/apertium-eng_lanc/eng_lanc.automorf.bin'
>> (No such file or directory)
>> at java.io.RandomAccessFile.open(Native Method)
>> at java.io.RandomAccessFile.<init>(RandomAccessFile.java:241)
>> at org.apertium.utils.IOUtils.memmap(IOUtils.java:495)
>> at org.apertium.utils.IOUtils.openFileAsByteBuffer(IOUtils.java
>> :322)
>> at org.apertium.lttoolbox.LTProc.doMain(LTProc.java:224)
>> at org.apertium.pipeline.Dispatcher.doLTProc(Dispatcher.java:180)
>> at org.apertium.pipeline.Dispatcher.dispatch(Dispatcher.java:264)
>> at apertiumview.Pipeline$PipelineTask.run(Pipeline.java:129)
>>
>> 2. Try the C++ option
>> New error:
>>
>> Return value: 1
>> Error: Cannot not open file
>> ''/home/rob/Deployed/apertium/eng_lanc/apertium-eng_lanc/eng
>> _lanc.automorf.bin''.
>> at apertiumview.Pipeline$1.run(Pipeline.java:40)
>>
>> Error repeated for both permission sets.
>>
>> 3. Try online option, for temporary download of pairs
>> Yes, that works.
>>
>>
>> == NOTES
>> I initially thought the problem was file permissions. However, the
>> exception stacktrace suggests the problem is in, or starts in, the
>> library 'lttools' bundled with apertium-viewer, not apertium-viewer
>> itself. Apertium viewer imports this library, then the error is thrown
>> when the apertium-viewer tries to call via 'Dispatch'.
>>
>> A quick look at the source code for the library,
>>
>> https://svn.code.sf.net/p/apertium/svn/trunk/lttoolbox-java/
>> src/org/apertium/utils/IOUtils.java
>>
>> shows me typically extensive, fairly standard Java file-load routines.
>> But, if I had the computer, I would start there?
>>
>>
>> Hope this is of some help (more help than I offered before)!
>> Robert
>>
>>
>>
>> On Mon, 2017-01-16 at 11:05 +0100, Jacob Nordfalk wrote:
>> > Saluton Donald,
>> >
>> >
>> > Pardonu pro mia malfrua respondo.
>> >
>> >
>> > First, it seems that the easy to run Apertium-viewer and the easy
>> > precompiled online pairs was what got you hooked on Apertium and gave
>> > you an insight on how the system work ('a way to show each step as
>> > it's done rather than just the final results. ')
>> >
>> >
>> > I'm very happy for that, as that is one of the primary goals
>> > of Apertium-viewer :-)
>> >
>> >
>> >
>> >
>> >
>> > The other primary goal of Apertium-viewer is to be an accessible/easy
>> > but powerfull tool to develop language pairs.
>> > To do that (be a language pair developer) you have to download and
>> > compile pairs from source (SVN), as well as the C++ libraries which
>> > Apertium consists of.
>> >
>> > Apertium-viewer can automate some steps but it is not a tool that sets
>> > up a complete develeopment environment for you.
>> >
>> >
>> >
>> > So, your message indeed does a lot of good, it warms my heart and you
>> > have got plenty of feedback on next step - which is to get your basic
>> > Apertium developer C++ tools working.
>> >
>> >
>> > I wont have time in the next month or so but I'd love to repair bugs
>> > in Apertium-viewer, so everyone (especially Robert) please keep the
>> > tool around so we can get the bugs fixed.
>> >
>> >
>> > Note that Apertium-viewer has 2 ways of working - using a buildin Java
>> > version or invoking an external C++ version. Try using the C++
>> > version if you get errors in the Java version:
>> > Billede indsat i tekst 1
>> >
>> >
>> >
>> >
>> >
>> > Yours,
>> > Jacob
>> >
>> >
>> >
>> >
>> >
>> > 2017-01-16 10:29 GMT+01:00 robert <[email protected]>:
>> > Hello Donald,
>> >
>> > - "not spend years or decades trying to figure out what I need
>> > to know
>> > in order to begin"
>> > I felt the same way. I did not know if Apertium would do what
>> > I wanted.
>> > I gave myself a week of evenings to see *something*. Then, I
>> > said to
>> > myself, I will abandon my efforts, as I could not afford the
>> > time.
>> >
>> > (...and now I am, when I am able, working on the Wiki, to
>> > contribute my
>> > experience back)
>> >
>> >
>> > - "adding a way to show each step as it's done rather than
>> > just the
>> > final results"
>> > For me, it was a breakthrough when I understood the 'modes',
>> > and how the
>> > effort put into templates, mode generation, and release
>> > software has
>> > made debugging and building new pairs fluent. I can understand
>> > your
>> > frustration.
>> >
>> > I guess what you need, above all, right now, is a fluent
>> > workflow? I'm
>> > not someone who demands sheets of configuration information
>> > before
>> > telling someone to read the manual, but...
>> >
>> > It may be helpful for the list (you may have explained on
>> > IRC?) if we
>> > knew the general environment you are working in, or are
>> > prepared to work
>> > in? Windows/Linux/Mac? My solution was clearly Linux-based
>> > (after the
>> > Java tools threw errors that I was not prepared to fight
>> > with).
>> >
>> >
>> > - "someone who knows what they're doing"
>> > I doubt anyone would claim to understand or recall the whole
>> > scope of
>> > Apertium? Some of the orphaned projects are remarkable and
>> > bizarre. Only
>> > a few says ago, I was swapping information about film
>> > subtitling---I was
>> > told this has not been tried in a long time, yet there is a
>> > lot of
>> > software in the repository.
>> >
>> > Notes: The 'documentation' front page remark about the manual
>> > is not
>> > fully correct. Important: for disambiguation, the project is
>> > now leaning
>> > towards the 'lexical selection' module. Perhaps you have good
>> > reason for
>> > using HFST?
>> >
>> > Adding to the links suggested by Hèctor... Ok, I understand
>> > your main
>> > problem is the workflow. But until you have one, or as a side
>> > interest
>> > while you/we find a way forward, here is my proposed reference
>> > page for
>> > the stages/modules in Apertium. This is intended to support
>> > the new
>> > 'stages' diagram, and contains a huge mass of information,
>> > links, and
>> > examples,
>> >
>> > http://wiki.apertium.org/wiki/User:Rcrowther
>> >
>> > It is difficult to read and understand, but I know I would
>> > have welcomed
>> > finding this page when I first came to Apertium.
>> >
>> > Please do not depend on this page staying at this link. I
>> > would like to
>> > publish to the Wiki.
>> >
>> > Regards,
>> > Rob
>> >
>> >
>> > On Sun, 2017-01-15 at 07:51 -0800, Donald Arthur Kronos wrote:
>> > > Thanks Robert. Yes, what I had actually originally been
>> > asking about
>> > > was the possibility of someone who knows what they're doing,
>> > adding a
>> > > way to show each step as it's done rather than just the
>> > final results.
>> > > I'm a bit of a newbie when it comes to Linux, so I expect it
>> > would
>> > > take quite a while for me to accomplish such a task on my
>> > own... so I
>> > > asked, because what I want to do is get working on making
>> > language
>> > > pairs as quickly as possible, not spend years or decades
>> > trying to
>> > > figure out what I need to know in order to begin.
>> > >
>> > >
>> > > So far, I have been working on experimental language pairs
>> > using HFST
>> > > on one or both of the monolingual resources, and there are
>> > many things
>> > > which are still too mysterious for me to make the progress I
>> > am trying
>> > > to.
>> > >
>> > >
>> > > Donald A. Kronos
>> > >
>> > >
>> > >
>> > > On Sun, Jan 15, 2017 at 7:27 AM, robert
>> > <[email protected]>
>> > > wrote:
>> > > Hello Donald,
>> > >
>> > > I do not know if my experience is relevant, but I
>> > started a
>> > > new language
>> > > pair two months ago. I had no knowledge of Apertium.
>> > I should
>> > > add also
>> > > that I am not a trained linguist, nor a practising
>> > translator.
>> > > My work
>> > > has been (at best) 'nursery' level. However, it has
>> > been
>> > > successful for
>> > > my own ends.
>> > >
>> > > I do not want to talk about your problems directly,
>> > as Kevin
>> > > has
>> > > answered directly about one issue, and we could
>> > spend a lot of
>> > > time
>> > > hunting down problems.
>> > >
>> > > 'apertium-viewer' is the program I think people
>> > mean. It shows
>> > > the
>> > > various steps in translation. It is a Java program,
>> > and I was
>> > > unable to
>> > > get any Apertium Java program working; I've seen the
>> > message
>> > > about
>> > > 'apertium-xxx-yyy.automorf' several times. I use
>> > Linux with an
>> > > OpenJDK,
>> > > and was very reluctant to try a full Oracle install
>> > (an
>> > > obvious next
>> > > step).
>> > >
>> > > However, the Apertium project is unlike many
>> > projects I have
>> > > visited. It
>> > > seems to have been assembled from bursts of
>> > enthusiasm, with a
>> > > great
>> > > deal of love, and much input from professional
>> > thought. On a
>> > > relative
>> > > scale, there is more code lying round this project
>> > than any
>> > > other I have
>> > > encountered.
>> > >
>> > > My solution was my solution, and may not work for
>> > you. I am
>> > > not
>> > > recommending this approach, but using it as an
>> > illustration. I
>> > > used
>> > > 'apertium-tolk', which I think is deprecated. But it
>> > is in
>> > > Linux
>> > > repositories and, a Python program, it is still
>> > working. Once
>> > > I had
>> > > apertium-tolk working, I could build a basic new
>> > pair and see
>> > > results.
>> > >
>> > > Then I needed, as you have suggested you need, to
>> > see the
>> > > steps in
>> > > translation. For this, since I am working in Linux,
>> > I was able
>> > > to use
>> > > the commandline and modes. Please see this Wiki
>> > page,
>> > >
>> > > http://wiki.apertium.org/wiki/Modes_introduction
>> > >
>> > > I wrote this page only three weeks(?) ago. I do not
>> > know how
>> > > far
>> > > advanced you are in understanding, but here is my
>> > proposed
>> > > replacement
>> > > for the overall diagram of Apertium structure,
>> > >
>> > >
>> > http://wiki.apertium.org/wiki/File:Apertium-structure.svg
>> > >
>> > > I am not recommending these wiki contributions
>> > because I
>> > > authored them.
>> > > When I developed my pair, I became very fond of the
>> > Apertium
>> > > Wiki, due
>> > > to the generous efforts of many contributors, and
>> > the
>> > > astonishing range
>> > > of material. But I have come to realise that there
>> > is a layer
>> > > of
>> > > material missing; of general introductions and
>> > indexing of
>> > > material. If
>> > > you have any ideas about where I could work, please
>> > tell me
>> > > (yes, I have
>> > > written Wiki pages to fully update installation and
>> > new pair
>> > > instructions).
>> > >
>> > > And I would add, my experience with Apertium, though
>> > > occasionally
>> > > frustrating, has lead to results which have been
>> > wonderful. I
>> > > only wish,
>> > > and have been actively seeking, time and
>> > justification to
>> > > continue. I
>> > > hope the problems you have encountered do not spoil
>> > your
>> > > projects.
>> > >
>> > > Regards,
>> > > Rob
>> > >
>> > > On Sat, 2017-01-14 at 15:12 -0800, Donald Arthur
>> > Kronos wrote:
>> > > > I do not know the proper procedures for a "mailing
>> > list"
>> > > such as this,
>> > > > but I'm trying very hard to learn enough of how
>> > Apertium
>> > > works to
>> > > > develop language pairs on it, so I'm giving this a
>> > try.
>> > > >
>> > > >
>> > > > I had been asking on the #apertium IRC channel
>> > about a way
>> > > to see the
>> > > > steps being taken by Apertium in the process of
>> > translating
>> > > something,
>> > > > especially in an experimental language pair that
>> > one hasn't
>> > > shared
>> > > > with the public yet (in the early stages of
>> > learning how to
>> > > make
>> > > > monolingual resources and language pairs), and
>> > perhaps even
>> > > the
>> > > > potential addition of step-by-step round-trip
>> > translation
>> > > support,
>> > > > from a source language to a target language and
>> > back, seeing
>> > > the steps
>> > > > along the way, and I was directed to Apertium-view
>> > as a tool
>> > > which
>> > > > should be able to do at least part of what I was
>> > looking
>> > > for.
>> > > >
>> > > >
>> > > > After downloading Apertium-view and getting it to
>> > work on an
>> > > > online-language pair I could see that it would
>> > serve most of
>> > > the
>> > > > purpose I had in mind, and even includes
>> > substantial
>> > > additional
>> > > > functionality, but unfortunately when I tried it
>> > with a
>> > > local
>> > > > experimental language pair it complained that the
>> > mode file
>> > > I
>> > > > specified could not be found. So, I made a new
>> > experimental
>> > > language
>> > > > pair on my system's main drive, thinking perhaps
>> > it can't
>> > > see my
>> > > > removable media, and that got it to see the mode
>> > file, but
>> > > then it
>> > > > claims all the other files it needs cannot be
>> > found.
>> > > Actually, there
>> > > > were more steps involved than what I'm describing,
>> > but I'm
>> > > trying to
>> > > > keep this simple. For example, I noticed the
>> > original error
>> > > message
>> > > > gave the address of the mode file twice, in one
>> > correctly
>> > > case
>> > > > starting with "/" and in the other case not, so...
>> > yeah...
>> > > long story.
>> > > > Anyway, I have been trying to figure out how to
>> > contact the
>> > > developer
>> > > > of Apertium-view and here's a copy of the text of
>> > the e-mail
>> > > I
>> > > > recently sent out as one attempt...
>> > > >
>> > > >
>> > > > "Saluton. I hope this reaches the right person.
>> > > >
>> > > >
>> > > > I am trying to work on language pairs on Apertium,
>> > but am
>> > > having
>> > > > trouble finding the information I need, so I
>> > recently asked
>> > > about any
>> > > > tool that could help me see the steps Apertium is
>> > taking as
>> > > it
>> > > > translates from one language to another, and I was
>> > directed
>> > > to
>> > > > Apertium-viewer. However, I have been unable to
>> > get it to
>> > > work for my
>> > > > local experimental language pairs. I see in the
>> > wiki page
>> > > that it is
>> > > > supposed to work with SVN source directories but I
>> > still
>> > > have no clue
>> > > > how to set up such a thing, nor do I think my
>> > experimental
>> > > language
>> > > > pairs are ready for that stage since I am still
>> > trying to
>> > > learn how to
>> > > > make them properly. Apertium-view cannot seem to
>> > open any
>> > > files at all
>> > > > from my removable media (which is the only place I
>> > really
>> > > have space
>> > > > to work) and even when I made an experimental
>> > language pair
>> > > in my home
>> > > > directory and was able to get it to open a mode
>> > file, it
>> > > still failed
>> > > > to be able to open any of the files it then needed
>> > to
>> > > proceed. For
>> > > > example, it says that it cannot open
>> > > > "/home/tz/apertium-qlt-qay.automorf.bin" even
>> > though that
>> > > file does
>> > > > exist, is not on removable media, and has its
>> > permissions
>> > > set so that
>> > > > it should be accessible. I'm somewhat of a newbie
>> > on Linux
>> > > so it MAY
>> > > > BE something I'm doing wrong, but I have no way of
>> > knowing.
>> > > Please,
>> > > > help if you can."
>> > > >
>> > > >
>> > > > ---
>> > > >
>> > > >
>> > > > Maybe he'll see this... or maybe someone else will
>> > see it
>> > > and be able
>> > > > to tell me how to work around it, or perhaps my
>> > expressed
>> > > intent will
>> > > > inspire something I hadn't even directly thought
>> > of.
>> > > Regardless, I
>> > > > hope this message does some good, somehow.
>> > > >
>> > > >
>> > > > Donald Arthur Kronos
>> > >
>> > > >
>> > >
>> > ------------------------------------------------------------
>> ------------------
>> > > > Developer Access Program for Intel Xeon Phi
>> > Processors
>> > > > Access to Intel Xeon Phi processor-based developer
>> > > platforms.
>> > > > With one year of Intel Parallel Studio XE.
>> > > > Training and support from Colfax.
>> > > > Order your platform today. http://sdm.link/xeonphi
>> > > > _______________________________________________
>> > > > Apertium-stuff mailing list
>> > > > [email protected]
>> > > >
>> > https://lists.sourceforge.net/lists/listinfo/apertium-stuff
>> > >
>> > >
>> > >
>> > >
>> > ------------------------------------------------------------
>> ------------------
>> > > Developer Access Program for Intel Xeon Phi
>> > Processors
>> > > Access to Intel Xeon Phi processor-based developer
>> > platforms.
>> > > With one year of Intel Parallel Studio XE.
>> > > Training and support from Colfax.
>> > > Order your platform today. http://sdm.link/xeonphi
>> > > _______________________________________________
>> > > Apertium-stuff mailing list
>> > > [email protected]
>> > >
>> > https://lists.sourceforge.net/lists/listinfo/apertium-stuff
>> > >
>> > >
>> > >
>> > -----------------------------------------------------------
>> -------------------
>> > > Developer Access Program for Intel Xeon Phi Processors
>> > > Access to Intel Xeon Phi processor-based developer
>> > platforms.
>> > > With one year of Intel Parallel Studio XE.
>> > > Training and support from Colfax.
>> > > Order your platform today. http://sdm.link/xeonphi
>> > > _______________________________________________
>> > > Apertium-stuff mailing list
>> > > [email protected]
>> > > https://lists.sourceforge.net/lists/listinfo/apertium-stuff
>> >
>> >
>> >
>> > -----------------------------------------------------------
>> -------------------
>> > Developer Access Program for Intel Xeon Phi Processors
>> > Access to Intel Xeon Phi processor-based developer platforms.
>> > With one year of Intel Parallel Studio XE.
>> > Training and support from Colfax.
>> > Order your platform today. http://sdm.link/xeonphi
>> > _______________________________________________
>> > Apertium-stuff mailing list
>> > [email protected]
>> > https://lists.sourceforge.net/lists/listinfo/apertium-stuff
>> >
>> >
>> >
>> >
>> >
>> > --
>> > Jacob Nordfalk
>> > javabog.dk
>> >
>> > Androidudvikler og -underviser på DTU og Lund&Bendsen
>> >
>> > ------------------------------------------------------------
>> ------------------
>> > Developer Access Program for Intel Xeon Phi Processors
>> > Access to Intel Xeon Phi processor-based developer platforms.
>> > With one year of Intel Parallel Studio XE.
>> > Training and support from Colfax.
>> > Order your platform today. http://sdm.link/xeonphi
>> > _______________________________________________
>> > Apertium-stuff mailing list
>> > [email protected]
>> > https://lists.sourceforge.net/lists/listinfo/apertium-stuff
>>
>>
>>
>
>
> --
> Jacob Nordfalk <http://profiles.google.com/jacob.nordfalk>
> javabog.dk
> Androidudvikler og -underviser på DTU
> <http://www.cv.diplom.dtu.dk/Diplomuddannelser/IT-Diplomuddannelsen/Valgfrie-moduler/Mobilapplikationsudvikling/Formaal-og-laeringsmaal>
> og Lund&Bendsen
> <https://www.lundogbendsen.dk/undervisning/beskrivelse/LB1809/>
>
------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, SlashDot.org! http://sdm.link/slashdot
_______________________________________________
Apertium-stuff mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/apertium-stuff
