On Sep 10, 2008, at 12:30 PM, JQ Johnson wrote:
>> you have to be able to cope with the fact that the user may have
>> altered their registry - adding conflicting entries, removing
>> others, etc.
>
> Agreed. If rows of a table are independent, which they should be in a
> well-normalized SQL dat
> you have to be able to cope with the fact that the user may have
> altered their registry - adding conflicting entries, removing
> others, etc.
Agreed. If rows of a table are independent, which they should be in a
well-normalized SQL database, then one can add a column that indicates
th
Dorothea Salo wrote:
> Good point. Would I be right in hazarding a guess that this was
> originally designed as a gesture toward relatively simple updating of
> this set of configuration options (without direct database-mucking)?
> And that this is therefore a special case of the general issues wit
On Wed, Sep 10, 2008 at 10:36:54AM -0700, Mark Diggory wrote:
> The inclusion of new formats would have probably been best served as
> part of the SQL upgrade in
>
> http://dspace.svn.sf.net/svnroot/dspace/branches/dspace-1_5_x/dspace/
> etc/database_schema_14-15.sql
>
> Then it would have bee
On Sep 10, 2008, at 10:43 AM, Dorothea Salo wrote:
> On Wed, Sep 10, 2008 at 12:36 PM, Mark Diggory <[EMAIL PROTECTED]>
> wrote:
>> The inclusion of new formats would have probably been best served as
>> part of the SQL upgrade in
>>
>> http://dspace.svn.sf.net/svnroot/dspace/branches/dspace-1_
On Wed, Sep 10, 2008 at 12:36 PM, Mark Diggory <[EMAIL PROTECTED]> wrote:
> The inclusion of new formats would have probably been best served as
> part of the SQL upgrade in
>
> http://dspace.svn.sf.net/svnroot/dspace/branches/dspace-1_5_x/dspace/
> etc/database_schema_14-15.sql
>
> Then it would h
The inclusion of new formats would have probably been best served as
part of the SQL upgrade in
http://dspace.svn.sf.net/svnroot/dspace/branches/dspace-1_5_x/dspace/
etc/database_schema_14-15.sql
Then it would have been part of the update process. This XML file
format thing is a bit overly
Handling configuration files is a pretty well understood process and
there's good code in all of the modern open source package managers.
For example, if you use rpm then it compares MD5 checksums of the
ostensible original file, the current file, and the new file. See for
instance http:/
Going forward (and I say this without even looking at the code, mind
you) it would seem to make sense to keep the stock list that comes
with future cut versions separate from any additions/modifications,
providing an easy way for future upgrades to identify entries that
have been changed si
On Wed, Sep 10, 2008 at 10:29 AM, Mark H. Wood <[EMAIL PROTECTED]> wrote:
> Automatic updating is tricky.
Yes, it probably doesn't make a whole lot of sense as a development priority.
The most sensible solution might be to tuck additions to the stock
list into the upgrade instructions? As I see i
Automatic updating is tricky.
o We could empty the table and replace it from the new list. We
don't want that -- you lose any local additions.
o We could stuff in the stock definitions wholesale. You'd lose any
local modifications to stock types.
o We could walk the stock list and ins
Hi Dorothea,
the format registries are not updated as far as I know. The change only
applied to bitstream-format.xml which is read once on fresh install.
Sunny greetings
Claudia
Dorothea Salo schrieb:
> On Wed, Sep 10, 2008 at 7:16 AM, Claudia Jürgen
> <[EMAIL PROTECTED]> wrote:
>> Hi,
>>
>>
On Wed, Sep 10, 2008 at 7:16 AM, Claudia Jürgen
<[EMAIL PROTECTED]> wrote:
> Hi,
>
> just a quick note since March 2006 CSS is part of the default bitstream
> format registry,
Dumb question: does the DSpace upgrade process add lines to the
bitstream format registry, or do places that have been run
Hi,
just a quick note since March 2006 CSS is part of the default bitstream
format registry, the default entry is:
text/css
CSS
Cascading Style Sheets
1
false
css
Sunny greetings
Claudia
Dorothea Salo schrieb:
> On Tue, Sep 9, 2008 at 2:00 PM, Kyle Kal
On Tue, Sep 9, 2008 at 2:00 PM, Kyle Kaliebe <[EMAIL PROTECTED]> wrote:
> Does anyone know how to get the HTML bitstreams to correctly utilize the CSS
> bitstream? Thanks.
Speaking from been-there-done-that... the problem may be that CSS
isn't in the bitstream format registry, such that DSpace is
Hello All,
I am working on a batch load for one of our clients in which it has been
requested that a single directory website (all files with relative file
paths to the same directory) be loaded into a single item. I have
successfully loaded the files for the website and designated the index as
16 matches
Mail list logo