ofbiz-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Forrest Rae (JIRA)" <j...@apache.org>
Subject [jira] [Created] (OFBIZ-6770) createCustRequestContent Hasn't worked in 6 years and 7 months
Date Tue, 15 Dec 2015 06:02:46 GMT
Forrest Rae created OFBIZ-6770:
----------------------------------

             Summary: createCustRequestContent Hasn't worked in 6 years and 7 months
                 Key: OFBIZ-6770
                 URL: https://issues.apache.org/jira/browse/OFBIZ-6770
             Project: OFBiz
          Issue Type: Bug
          Components: order
    Affects Versions: Trunk, Release Branch 14.12, Release Branch 13.07, Release Branch 12.04,
Release Branch 11.04
            Reporter: Forrest Rae


This commit disabled the mimeTypeId argument in applications/order/widget/ordermgr/CustRequestForms.xml,
in order to provide "various fixes"

https://fisheye6.atlassian.com/changelog/ofbiz?cs=768661

I wonder how many people have been frustrated by this when looking for example code on how
to upload documents.

This patch isn't really a fix, but more of a work around.  The issue lies in the seemly arbitrary
check of the mimeTypeId parameter in applications/order/script/org/ofbiz/order/request/CustRequestEvents.xml,
lines 52-54.

First if-compare is "Does the web browser's choice of mimeTypeId equal what the user thinks
the mimeTypeId is?"

Second if-compare is "Did the user select anything at all in the mimeTypeId dropdown?"

Third choice is "Did the user submit mimeTypeId form field, but remove the value, submitting
a blank form field?"

My question is why are we even doing any of this when the event just goes on to set the mimeType
to what the web browser decided it was, and then to call createContentFromUploadedFile service?
 Also, if mimeTypeId is in fact empty, the call to createContentFromUploadedFile calls createDataResource,
which at applications/content/script/org/ofbiz/content/data/DataServices.xml:53 calls org.apache.tika.Tika#detect(byte[])
to detect the mimeType anyway.  Why not just let tika handle the mimeType?

Anyway, I dunno enough about the Content component to make the decision here.  I've attached
three different patches.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message