- Zope - Zope lists

Re: [Zope] DTML batch processing variables
by Martijn Pieters 25 Jun '99

25 Jun '99

At 13:55 25/06/99 , Oleg Broytmann wrote: >Hi! > >On Fri, 25 Jun 1999, Martijn Pieters wrote: > > >next-sequence-size is alway "1", and previous-sequence-size is > > >"-90","-70","-50","-30","-10" everything else works as described. > > > > What version of Zope and Python are you using? There is a known bug with > > all versions of Zope up to Zope 2.0.0a2 (fixed in Zope 2.0.0a3) that > > becomes effective when using python 1.5.2. > > I wonder how is it possible to develop a program that does different >calculations (very simle math, I think) with 1.5.1 and 1.5.2? > > Any cahnce we'd see a patch? I'd like to incorporate it into 1.10.3... It has nothing to do with simple math, but everything with the interpretation of None compaired to 0's. Something that was implemented wrongy in python 1.5.1, and fixed in 1.5.2. This change brought out a bug in the code, which hitherto had managed to miraculously work anyway. 1.10.3 is meant to run with python 1.5.1, not 1.5.2, and has only been tested with 1.5.1. There are other problems concerning running Zope 1.10.3 with version 1.5.1 than this one, like Netscape Publishing, FTP uploads and WebDAV being broken. If you insist, here is a context diff from the CVS you could apply, it probably has slightly differing line numbers, but you should be able to apply it to DT_In.py anyway, as patch will correct for this. It basically replaces all occurrences of None within calls to the method 'opt' with 0: *************** *** 531,537 **** try: if previous: if first > 0: ! pstart,pend,psize=opt(None,first+overlap, sz,orphan,sequence) kw['previous-sequence']=1 kw['previous-sequence-start-index']=pstart-1 --- 531,537 ---- try: if previous: if first > 0: ! pstart,pend,psize=opt(0,first+overlap, sz,orphan,sequence) kw['previous-sequence']=1 kw['previous-sequence-start-index']=pstart-1 *************** *** 547,553 **** # there are more items, without actually # computing a length: sequence[end] ! pstart,pend,psize=opt(end+1-overlap,None, sz,orphan,sequence) kw['next-sequence']=1 kw['next-sequence-start-index']=pstart-1 --- 547,553 ---- # there are more items, without actually # computing a length: sequence[end] ! pstart,pend,psize=opt(end+1-overlap,0, sz,orphan,sequence) kw['next-sequence']=1 kw['next-sequence-start-index']=pstart-1 *************** *** 563,569 **** validate=md.validate for index in range(first,end): if index==first and index > 0: ! pstart,pend,psize=opt(None,index+overlap, sz,orphan,sequence) kw['previous-sequence']=1 kw['previous-sequence-start-index']=pstart-1 --- 563,569 ---- validate=md.validate for index in range(first,end): if index==first and index > 0: ! pstart,pend,psize=opt(0,index+overlap, sz,orphan,sequence) kw['previous-sequence']=1 kw['previous-sequence-start-index']=pstart-1 *************** *** 577,583 **** # test whether there are more items, # without actually computing a length: sequence[end] ! pstart,pend,psize=opt(end+1-overlap,None, sz,orphan,sequence) kw['previous-sequence']=0 kw['next-sequence']=1 --- 577,583 ---- # test whether there are more items, # without actually computing a length: sequence[end] ! pstart,pend,psize=opt(end+1-overlap,0, sz,orphan,sequence) kw['previous-sequence']=0 kw['next-sequence']=1 -- Martijn Pieters, Web Developer | Antraciet http://www.antraciet.nl | Tel: +31-35-7502100 Fax: +31-35-7502111 | mailto:mj@antraciet.nl http://www.antraciet.nl/~mj | PGP: http://wwwkeys.nl.pgp.net:11371/pks/lookup?op=get&search=0xA8A32149 ------------------------------------------

3 2

RE: [Zope-dev] DateTime problem
by Brian Lloyd 25 Jun '99

25 Jun '99

> Hi all, > > I've fixed this problem as per last week and posted a patch on the > Collector. To grab this patch plus patches for: > * Database corruption under windows > * Import typo > * BoboPOS getting in the way of Zodb3 > > please goto: > http://student.uq.edu.au/~s341625 > > note that because I haven't figured out diff/patch under > windows yet you > may have to apply the context sensitive patches yourself ;) > > PS How long before stuff goes through the collector? > > Cheers, > > Anthony Pfrunder Thanks for sending it in. We haven't forgotten it- there some other issues to deal with as well, so it's just a matter of having the time to get to everything... :) Brian Lloyd brian(a)digicool.com Software Engineer 540.371.6909 Digital Creations http://www.digicool.com

1 0

RE: [Zope] Tree tag and SQL methods?
by Alexander Staubo 25 Jun '99

25 Jun '99

I haven't tested this yet, but it seems sound; I'll let you know if it doesn't. :-) Now perhaps you can assist in another similar case with the #tree tag. Basically I have a two-level hierarchy of (simplified) articles and their home publication. The article database might be something like this: Title: Publication: Foo Bar Mag Bar Fnargh Times ... I seek to render a "tree" where the publication is the tree node, and the list of articles (by their title) are listed under each node. A little like the "Quick Start" thing. Click on the "plus" icon beside the publication and you get a (non-tree) list of articles. Actually I get this is, strictly speaking, a one-level hierarchy. While this structure is simple enough that could stay clear of using the #tree tag, I'd still like to try. I tried:    thinking that Query() would return the publications, and each publication would use the DTML Method Leaf to render its contents, but no. What am I missing? -- Alexander Staubo http://www.mop.no/~alex/ "What the hell, he thought, you're only young once, and threw himself out of the window. That would at least keep the element of surprise on his side." --Douglas Adams, _The Hitchhiker's Guide to the Galaxy_ >-----Original Message----- >From: Phillip J. Eby [mailto:pje@telecommunity.com] >Sent: 25. juni 1999 18:38 >To: Alexander Staubo; Zope Mailing List (E-mail) >Subject: RE: [Zope] Tree tag and SQL methods? > > >I looked back at what I posted, then at what I actually did in >code, and >realized why what I told you is wrong. > >See, here's what the tricky part of the tree tag is. It wants as its >name/expr a single object to be the root; it doesn't grok the >concept of >multiple roots. So try it like this: > > > >... > > > >Thus, the first query looks for roots, and subsequent queries >should have >obj_id filled in from the prior queries' results. Note that >query() must >look for root entries if obj_id is 0. As you can see, there >are two minor >changes from what I said before - first, there's no need to have the >expr="query()" that I told you, it only messes things up. :) >Second, a >namespace is needed to provide a default value on the first >execution of >the query. > > >At 02:39 AM 6/25/99 +0200, Alexander Staubo wrote: >>[This post is old, but I haven't ended up tackling the topic >until now.] >> >>>-----Original Message----- >>>From: zope-admin(a)zope.org [mailto:zope-admin@zope.org]On Behalf Of >>>Phillip J. Eby >>>Sent: 9. juni 1999 21:21 >>>To: Alexander Staubo; Zope Mailing List (E-mail) >>>Subject: Re: [Zope] Tree tag and SQL methods? >>> >>>No need. Just include the id as a field in your result, and >have your >>>branches_expr pass that id back into the query. Your query >>>can then look >>>like: >>> >>> >>>SELECT * FROM data WHERE >>> >>>parent_id= >>> >>>parent_id IS NULL >>> >>> >>> >>>And your tree would look like: >>> >>> >>>... etc. >>> >>>It's been a long while since I did this, so I may be a bit off >>>here. (It's >>>been so long in fact, that when I did it, there was no such >thing as a >>>#sqlvar tag, and tree didn't *have* a branches_expr option! >Which made >>>this much more difficult to do, especially with pluggable >brains thrown >>>into the mix.) >> >>Nope, doesn't work. First of all, #tree doesn't seem to handle SQL >>Method output the way I want -- I don't know why. Secondly, the >>branches_expr expression doesn't compute because obj_id isn't a known >>variable at this point; it should be written as >> >> >> >>But that doesn't seem to work either. Testing with a simple >record with >>one parent record, #tree does render one item, but I can't get at the >>SQL row: For example,  inside the #tree tag >gives me a >>NameError exception. >> >>So how do I do it? >> >>-- >>Alexander Staubo http://www.mop.no/~alex/ >>"What the hell, he thought, you're only young once, and threw >>himself out of the window. That would at least keep the element of >>surprise on his side." >>--Douglas Adams, _The Hitchhiker's Guide to the Galaxy_ >> >> >>_______________________________________________ >>Zope maillist - Zope(a)zope.org >>http://www.zope.org/mailman/listinfo/zope >> >>(For developer-specific issues, use the companion list, >>zope-dev(a)zope.org - http://www.zope.org/mailman/listinfo/zope-dev ) > >

2 1

Re: [Zope] DTML batch processing variables
by Joachim Schmitz 25 Jun '99

25 Jun '99

Hi Martijn, thanks for the patch, it solved the problem. On Fri, 25 Jun 1999, Martijn Pieters wrote: > > It has nothing to do with simple math, but everything with the > interpretation of None compared to 0's. Something that was implemented > wrongly in python 1.5.1, and fixed in 1.5.2. This change brought out a bug > in the code, which hitherto had managed to miraculously work anyway. > > 1.10.3 is meant to run with python 1.5.1, not 1.5.2, and has only been > tested with 1.5.1. There are other problems concerning running Zope 1.10.3 > with version 1.5.1 than this one, like Netscape Publishing, FTP uploads and > WebDAV being broken. > > If you insist, here is a context diff from the CVS you could apply, it > probably has slightly differing line numbers, but you should be able to > apply it to DT_In.py anyway, as patch will correct for this. It basically > replaces all occurrences of None within calls to the method 'opt' with 0: > Gruß Joachim Schmitz

1 0

RE: [Zope] Tree tag and SQL methods?
by Alexander Staubo 25 Jun '99

25 Jun '99

[This post is old, but I haven't ended up tackling the topic until now.] >-----Original Message----- >From: zope-admin(a)zope.org [mailto:zope-admin@zope.org]On Behalf Of >Phillip J. Eby >Sent: 9. juni 1999 21:21 >To: Alexander Staubo; Zope Mailing List (E-mail) >Subject: Re: [Zope] Tree tag and SQL methods? > > >At 07:08 PM 6/9/99 +0200, Alexander Staubo wrote: >>I don't have a head for figuring this out today, but maybe >somebody here >>can help me out. >> >>I'm thinking of exploiting the #tree tag in conjunction with >one or more >>SQL method. Say I have a hierarchical-ish table with id/parent_id >>references, and I'd like to display this information in a nice >>collapsible tree. Possible? Possibly the expr= or branches_expr= >>arguments to #tree could evaluate the result of an SQL >method, although >>it sounds to me like this would require one query per tree level. >>Possibly you could put the query results of the complete table in a >>temporary variable and evaluate on that? > >No need. Just include the id as a field in your result, and have your >branches_expr pass that id back into the query. Your query >can then look >like: > > >SELECT * FROM data WHERE > >parent_id= > >parent_id IS NULL > > > >And your tree would look like: > > >... etc. > >It's been a long while since I did this, so I may be a bit off >here. (It's >been so long in fact, that when I did it, there was no such thing as a >#sqlvar tag, and tree didn't *have* a branches_expr option! Which made >this much more difficult to do, especially with pluggable brains thrown >into the mix.) Nope, doesn't work. First of all, #tree doesn't seem to handle SQL Method output the way I want -- I don't know why. Secondly, the branches_expr expression doesn't compute because obj_id isn't a known variable at this point; it should be written as  But that doesn't seem to work either. Testing with a simple record with one parent record, #tree does render one item, but I can't get at the SQL row: For example,  inside the #tree tag gives me a NameError exception. So how do I do it? -- Alexander Staubo http://www.mop.no/~alex/ "What the hell, he thought, you're only young once, and threw himself out of the window. That would at least keep the element of surprise on his side." --Douglas Adams, _The Hitchhiker's Guide to the Galaxy_

3 2

New version of mod_pcgi
by A.M. Kuchling 25 Jun '99

25 Jun '99

I've finally gotten around to fixing the most serious outstanding problems with mod_pcgi; it is now actually usable. PATH_INFO is now set correctly, so Zope creates URL references that are reasonable, and the Authorization HTTP header is passed along so that you can actually log in. In my limited testing, everything I tried in Zope seemed to work fine using mod_pcgi. In short, you could actually start using it for real now. The code is in the usual place, ftp://starship.python.net/pub/crew/amk/new/ Please download it and try it out; I'm going to work on some Web pages and documentation for it. -- A.M. Kuchling http://starship.python.net/crew/amk/ Well, you are the best I have, Cluracan, and to say that is to say little indeed. -- Queen Titania, in SANDMAN #52: "Cluracan's Tale"

6 10

DTML batch processing variables
by Joachim Schmitz 25 Jun '99

25 Jun '99

Hi, I tried the example from the DTML userguide with my query and get wrong values for: next-sequence-size and previous-sequence-size, here is my dtml-example:        <a href="qs="> (vorherige  Adressen )</a>    E-Mail.DE hat insgesamt  Adresse n gefunden.  <H4>.   </H4> , ;   Tel:  Fax:   <a HREF="mailto:"></a> geschäftlich   <a HREF="mailto:"></a> privat         <a href="qs="> (die nächsten  Adressen )</a>   Es wurden leider keine Einträge gefunden  When for example 111 records are found: next-sequence-size is alway "1", and previous-sequence-size is "-90","-70","-50","-30","-10" everything else works as described. Gruß Joachim Schmitz

4 4

CVS checkout errors?
by Alexander Staubo 25 Jun '99

25 Jun '99

List traffic is slow, so I thought I'd do my part in boosting the activity :-) Using CVS 1.9 I get lots of "internal errors" from CVS when checking out Zope. As far as I can determine, though, files are checked out completely. The error is: cvs checkout: internal error: repository string too short. Couldn't find anything in the CVS FAQ about this. -- Alexander Staubo http://www.mop.no/~alex/ "What the hell, he thought, you're only young once, and threw himself out of the window. That would at least keep the element of surprise on his side." --Douglas Adams, _The Hitchhiker's Guide to the Galaxy_

2 1

RE: [Zope] Startup error Zope 2.0.0a3: Error 139
by Alexander Staubo 25 Jun '99

25 Jun '99

I hear rumours that error code 139 indicates memory corruption. Also I just noticed the crash below has Python producing a core dump, anybody want to look at it? -- Alexander Staubo http://www.mop.no/~alex/ "What the hell, he thought, you're only young once, and threw himself out of the window. That would at least keep the element of surprise on his side." --Douglas Adams, _The Hitchhiker's Guide to the Galaxy_ >-----Original Message----- >From: Alexander Staubo [mailto:alex@mop.no] >Sent: 25. juni 1999 15:43 >To: Zope Mailing List (E-mail) >Subject: [Zope] Startup error Zope 2.0.0a3: Error 139 > > >I just installed Zope 2.0.0a3 (no CVS patches) on a Slackware Linux 2.2 >(libc6, Python 1.5.2) box. Of course, it refuses to start: > >firewall:/usr/local/zope-2.0.0a3# python z2.py -d 192.168.1.1 >-w 8080 -f >2121 >zdeamon: Fri Jun 25 15:32:33 1999: Houston, we have forked >zdeamon: Fri Jun 25 15:32:33 1999: Houston, we have forked >zdeamon: Fri Jun 25 15:32:33 1999: Hi, I just forked off a kid: 24905 >zdeamon: Fri Jun 25 15:32:35 1999: Aiieee! 24905 exited with >error code: >139 > >Curiously, no tracebacks? > >-- >Alexander Staubo http://www.mop.no/~alex/ >"What the hell, he thought, you're only young once, and threw >himself out of the window. That would at least keep the element of >surprise on his side." >--Douglas Adams, _The Hitchhiker's Guide to the Galaxy_ > > >_______________________________________________ >Zope maillist - Zope(a)zope.org >http://www.zope.org/mailman/listinfo/zope > >(For developer-specific issues, use the companion list, >zope-dev(a)zope.org - http://www.zope.org/mailman/listinfo/zope-dev ) >

1 0

Startup error Zope 2.0.0a3: Error 139
by Alexander Staubo 25 Jun '99

25 Jun '99

I just installed Zope 2.0.0a3 (no CVS patches) on a Slackware Linux 2.2 (libc6, Python 1.5.2) box. Of course, it refuses to start: firewall:/usr/local/zope-2.0.0a3# python z2.py -d 192.168.1.1 -w 8080 -f 2121 zdeamon: Fri Jun 25 15:32:33 1999: Houston, we have forked zdeamon: Fri Jun 25 15:32:33 1999: Houston, we have forked zdeamon: Fri Jun 25 15:32:33 1999: Hi, I just forked off a kid: 24905 zdeamon: Fri Jun 25 15:32:35 1999: Aiieee! 24905 exited with error code: 139 Curiously, no tracebacks? -- Alexander Staubo http://www.mop.no/~alex/ "What the hell, he thought, you're only young once, and threw himself out of the window. That would at least keep the element of surprise on his side." --Douglas Adams, _The Hitchhiker's Guide to the Galaxy_

1 0