Messages

46

Open Embedded / Building For The Sharp-based 'roms' With Oe

« on: October 25, 2006, 08:30:59 am »

For reasons I won't go into (in the interest of avoiding a flame war), I've gone back to the Cacko ROM and so I need to get a Sharp compatible dev env going again. I can always fall back on my DSL based system, but there are a lot of advantages to being able to use OpenEmbedded.

I know it's "theoretically" possible, but it seems the original page referenced by this thread is 404 and the only reference I can find on the OE wiki says:

Building packages for the Sharp ROM is as easy as setting

DISTRO = "sharprom-compatible"

in your local.conf and then compiling as usual. Unfortunately, the process is not yet 100% stable. So do not be surprised if you have some problems along the way.

Has anyone tried this recently, and if so could you summarize your results? The condition "not yet 100% stable" can be satisfied by 1% stable just as well as 99% stable. I'd love to know which end of that spectrum we're really talking about. Time is in short supply and I can't afford to waste a bunch on a doomed experiment.

47

Angstrom & OpenZaurus / Ir Problems With 3.5.4.1

« on: October 16, 2006, 12:38:37 pm »

let me tell the gnome, gentoo and kde folk that bugzilla doesn't work for users, so they can install something different, since you speak for *each and every* user.
[div align=\"right\"][a href=\"index.php?act=findpost&pid=144082\"][{POST_SNAPBACK}][/a][/div]

I didn't say don't use bugzilla, I said hide it behind a friendly interface.

Have you actually looked at KDE's bugzilla interface? They did exactly what I suggested and hid bugzilla's horrible bug entry form behind a nice user friendly wizard that presents only a couple fields at a time and actually explains what to do with them. Yeah, I still have to register an account, but since they don't plaster my registered email address all over the displayed bug reports, it's a minor inconvenience.

And no, I don't speak for all users. But I do have extensive experience providing support to paying customers. Since you're always pissed off about users not entering OE bug reports, I thought you might appreciate some suggestions that could help address that problem. I forgot that the only "acceptable" explanations are those that put the blame on the users.

OpenZaurus has such great potential, it's a shame it's shrouded in such an impenetrable cloud of arrogance.

48

Angstrom & OpenZaurus / Ir Problems With 3.5.4.1

« on: October 15, 2006, 11:16:34 pm »

Filed a bug about it?
[div align=\"right\"][a href=\"index.php?act=findpost&pid=144019\"][{POST_SNAPBACK}][/a][/div]

No, I haven't.

When I tried, my first problem was that the only choices for "product" were:

BitBake:    Bug reports for the BitBake build tool
Openembedded:     Queue for tracking tickets against the OE metadata repository and OE based distributions.

Neither of which seemed entirely appropriate. The OE Bugzilla DB has a "Zaurus Kernel" product which does seem like the correct place, but the submit page doesn't give me that option. I was a little worried that submitting incorrectly would get me labeled a "whiny Zaurus user" for whom the OE folks clearly have a serious dislike (and then what would my chances be of my bug ever being fixed).

But, not being one to give up easily, I decided to take a chance and pick Openembedded. Now I'm faced with a Bugzilla login screen. I need to register for an account and provide an email address. It has to be a working address because I can't log in until a password gets mailed to me. And then that address is going to get slapped onto my bug report with no obfuscation whatever making it a prime target for spammers. No thanks.

If you are serious about wanting end users to submit bug reports, then you need to make it easier and you need to provide at least a little protection for the submitter's personal information. And if you're really serious about it, then you need to hide that Bugzilla abomination on a developer's only site and give end users a nice simple "Submit a bug" form that doesn't require registration and doesn't have more than about 4 fields and a "submit" button. I should probably submit a bug report for that ... ERROR: RECURSIVE POST DETECTED

49

Angstrom & OpenZaurus / Ir Problems With 3.5.4.1

« on: October 14, 2006, 10:41:19 pm »

I have found a minor change on pxaficp_ir.c in kernel 2.6.18
[div align=\"right\"][a href=\"index.php?act=findpost&pid=143947\"][{POST_SNAPBACK}][/a][/div]

Nope. Problem is that pxaficp_ir is used only for the IRDA protocol. Yes, some devices can do TTY over IRDA (ircomm) but most keyboards are not among them.

Keyboards need IR support in the PXA serial port driver. Sharp had kludged a bunch of extra IR handling code into the 2.4 kernel serial port driver, but none of it is present in the 2.6 kernel serial driver. Someone with a lot of time on their hands is going to have to either figure out Sharp's (actually Lineo's) spaghetti code and port it to the 2.6 serial driver or they'll have to design in IR support from scratch. I have the knowledge but I expect to be working 60 hour weeks through Feb so not much time for volunteer kernel hacking. I'm actually looking to drop back to a 2.4 kernel based ROM so I can get my keyboard back.

50

Qt/Qtopia / Sharp Zaurus Sdk The Easy Way

« on: August 25, 2006, 07:52:20 am »

What's the su password? Is that dsl? I tried that but it said Authentication failure.
[div align=\"right\"][a href=\"index.php?act=findpost&pid=139747\"][{POST_SNAPBACK}][/a][/div]

DSL (like many distros) has root logins disabled by default. To get root, do a "sudo su" (no password is required unless you've set a password for the "dsl" user).

51

Zaurus - pdaXrom / Ir Keyboard Drivers - Alpha Release

« on: August 24, 2006, 01:55:52 pm »

But the problem (I guess) is with the map_shift array of the tokens, for example, when I press a key (with shift or caps mode), I get an scancode, it is ok, but when I look for the scancode in the map_shift it is 0 !!.

Ok, the first thing I should point out is don't trust my maps  I don't have a Palm keyboard so the maps in the driver came from me trying to reverse engineer the keymaps using the IRK driver source code and fuzzy photos of the keyboard scrounged from various websites.

and example is with the key of the "@" I got an scancode of 0x79, and looking at the array it is 0, but it is funny because just a line before there is the token ZK_AT, at the scancode 0x71, the same occur with other keys using shift or caps.

That just means that I screwed up  Change the 0 to ZK_AT and change the ZK_AT to 0.

The way things work is if the driver sees a scancode with the shift key down, it looks up that scancode in the "map_shift" array. If there's a token there, the driver sends it. If there's a 0 there, the drivers sends "shift down, token from map_normal, shift up". The Zaurus is a little odd in that respect. For some characters is expects a shift token followed by the regular key token, for others (like "@") it has special dedicated tokens that must be sent without the shift key.

I feel a little blind because I would like to use some keys but I do not know what token name match with that character.

Basically the only tokens that need to be in the "map_shift" table are those that are already there. You may just need to move them around to get them matched to the right keys.

Note that the "function" maps are similar. If the function key is down, the driver looks up the scancode in the map_fn (or whatever I called it). If there's a token there, the driver sends it. But if there's a zero there, the driver sends nothing.

52

Zaurus - pdaXrom / Ir Keyboard Drivers - Alpha Release

« on: August 22, 2006, 11:36:33 am »

but I changed it to: ZK_PERIODTMP = 64 ... ZK_PERIOD = 103

Only in that way I got a correct token link.
[div align=\"right\"][a href=\"index.php?act=findpost&pid=139401\"][{POST_SNAPBACK}][/a][/div]

Ok, the problem isn't in ztokens, it's in the scancode maps in the palm_3169ww.lua file. The way things work is that the palm_3169ww.lua file converts scancodes from the keyboard into a Zaurus keyboard scancode. When you hit "shift" on your kb, it's getting turned into ZK_PERIOD which is wrong. Changing ZK_PERIOD in ztokens just masks the problem and will break badly if you ever try to use a different keyboard.

What you need to do is edit palm_3169ww.lua and look at line 65. Note how the 7th entry on that line is "ZK_PERIOD". What that is telling zkbdd is that if it sees a 0x2e scancode, it should send the Zaurus the ZK_PERIOD scancode. Now, apparently on your keyboard it's "right shift" and not "period" that sends the 0x2e scancode. So you need to replace ZK_PERIOD on that line with ZK_SHIFTR. Similarly you need to find out what scancode your keyboard is sending when you press "period" and replace that entry in the "map_normal" list with ZK_PERIOD. You can see scancodes by running zkbdd in a terminal with the test9600 driver.

Once you get the hang of this, it becomes pretty easy to fix any misbehaving key without touching ztokens. Once you get it sorted out, send me a copy of your modified palm_3169ww.lua file and I'll incorporate it into the next release of zkbdd.

53

Python / Tkinter

« on: August 22, 2006, 06:34:18 am »

I don't know about Tkinter but Tk works fine under X/Qt in the Sharp ROM. Caveat: I've used Tk only via tclkit (which I mentioned to Bam on another thread). I am not sure about a direct install.

Felipe
[div align=\"right\"][a href=\"index.php?act=findpost&pid=139305\"][{POST_SNAPBACK}][/a][/div]

Yes, if you go down the X/Qt path, it may be theoretically possible. One would have to custom build the Tkinter stuff and I have no idea what that would entail. If you're writing new apps, it seems to me that just using PyQt would be a whole lot less work. PyQt is really an outstanding GUI framework ... far more sophisticated than Tkinter. If you're just trying to run an existing large Tkinter application, I'd be tempted to switch to OZ/GPE or pdaXrom.

54

Python / Tkinter

« on: August 21, 2006, 06:14:50 am »

I used tkinter from another feed to add to Kopsis python image but still get _tkinter not found)err rather not configured for your install) any ideas on how to get this to work?
[div align=\"right\"][a href=\"index.php?act=findpost&pid=139251\"][{POST_SNAPBACK}][/a][/div]

TkInter needs Tk which need X11. I don't believe you'll get it to work on a Qt based ROM. For Sharp/Sharp compatible ROMs, PyQt is pretty much your only choice for GUI Python apps.

55

Florida / What To Do In Orlando?

« on: August 20, 2006, 04:46:55 pm »

Well, you're only an hour or so from the Kennedy Space Center. You can easily spend up to a whole day there and see some really interesting and historic stuff. No one can truly appreciate the enormity of what the Apollo program acomplished until you've stood next to an actual Saturn V rocket!

56

Python / Loops And Such

« on: August 20, 2006, 08:10:36 am »

yea there was some issues, got it all solved, though not quite as nice as yours. I am new to c-style programming, mostly I diddle with visual basic(for my company work). I saw that python had a nice gui to play with pyQt so this "seemed" closest to vb, so I thought the learning curve would be a little more flat, but guess not. I will however need to start writing "functions" and "subs" in order to keep stuff organized and not spagetti code, not only that but reuseability will be nicer.
[div align=\"right\"][{POST_SNAPBACK}][/a][/div]

Congrats on getting your program working  Going from VB to any modern language is going to have bit of a learning curve, but if you stick with it you'll make yourself a much better programmer. However, I have to disagree with your statement that you're new to C style programming. What you call C style is technically referred to a "imperative" programming. Both VB and C (at least in conventional usage) are actuall very similar in abstract programming language terms.

What causes a lot of confusion for new Python programmers is that it offers a number of features borrowed from the world of "functional programming". Things like lists, first-class functions, list comprehensions and built-ins like map() and filter() let you break out of the imperative style when writing Python code.

Why would you want to do that? Well, research shows that for many types of problems functional programming techniques result in much smaller programs with far fewer bugs. It makes it possible to work with programs as mathematical expressions rather than cookbook recipes. Ask yourself "where are most of the bugs in my programs?" The answer is usually "loops and state". Note how my example contained neither  The function I defined wasn't there for traditional structured programming purity, it was there so I could pass it as a variable to the filter() built-in  The whole idea of passing functions as variables may sound a bit strange, but that's one of the keys to FP.

Now, just as one can write FORTRAN style code in C++, it's possible to write VB style code in Python. But if you really want to improve yourself as a programmer, focus on learning to use the FP features in Python. In the long run, you'll be glad you did  If you want to learn more [a href=\"http://www.defmacro.org/ramblings/fp.html]Functional Programming For The Rest Of Us[/url] is a good place to start.

57

Python / Loops And Such

« on: August 19, 2006, 10:14:53 pm »

I have this so far:

Code: [Select]

for item in dupes[]:
   mydupeitem=item
   for item in filter[]:
      if mydupeitem[2:] = item:
         pass
      else
         write......etc.[div align=\"right\"][a href=\"index.php?act=findpost&pid=139096\"][{POST_SNAPBACK}][/a][/div]

A few observations:

* Your "if mydupeitem..." line is using an assignment (=) instead of a compare (==).
* You're reusing your loop variable (item) in the inner loop. I'm not sure Python cares (I'm in the habit of not trying such things), but in other languages like C that would cause much breakage. Note: "item" is just a variable that gets set on each loop iteration. You can use any variable name there that you wish.
* I don't understand the empty [] brackets on the "for item ..." lines. They should cause syntax errors.

A more elegant solution (assuming you're using a recent version of Python would be something like:

Code: [Select]

def f(x):
    for item in filter:
        if x[2:] = item:
            return false
    return true

filtered_list = filter(f, dupes)
That will run f(x) on every element in dupes and set filtered_list equal to the list of those for which f(x) returned true (i.e. they did not match any items in the filter list).

58

Angstrom & OpenZaurus / Belkin F8u1500-e Keyboard

« on: August 19, 2006, 06:04:45 pm »

The problem is in the Kernel and it doesn't matter if you're running OPIE or GPE.

I've prototyped a 2.6 device driver that may do the trick. I was hoping to do some testing/debug this weekend, but it looks like I'll be spending most of the weekend doing automotive work. Perhaps I'll have some news next week.

59

New products and alternatives / Open Linux Mobile Phone By Trolltech

« on: August 15, 2006, 01:12:14 pm »

I'm not so sure that this "If we build it, they will come" strategy is going to accomplish much. Commercial developers aren't going to waste time on a platform with a path to the consumer that is speculative at best.

Open source developers may love it, but the first thing they're going to do is replace Trolltech's Qtopia with something else (OPIE or GPE) or spend all their time hacking in the kernel, not the application space. Not that there's anything wrong with that, it just doesn't help Trolltech with their goal of getting consumer friendly apps developed for the phone.

In the US, consumers in general don't even want smartphones. Of the few that do, most want little more than seamless desktop (Windows Outlook) integration, a good set of built-in PIM apps, and possibly Blackberry style email connectivity. Even with Windows Mobile devices (which are, for the most part, not crippled by the carriers), the fraction of users that ever install an app that didn't come bundled with the phone is surprisingly small. Now stir into that mix the schizophrenic (and corrupt) wireless carriers and it's pretty clear that smartphones will never be more than a tiny niche on this side of the pond.

I wish Trolltech luck. I'd really like to see Linux phones succeed, but my expectations are pretty low.

60

Angstrom & OpenZaurus / Bvdd Drivers Any Soon?

« on: August 12, 2006, 04:38:28 pm »

As anyone who's had to trudge through Sharp's 2.4 kernel hacks will tell you, you want Sharp to stay as far away from software development as possible  Now, if they were to offer some hands-off financial sponsorship to some of the existing Zaurus open source projects, that would really rock ... but I'm not going to hold my breath.