No. Maybe on Monday. john mcnally
Rodney Schneider wrote: > > Hi John, > > Any luck investigating this problem? > > Thanks, > > -- Rodney > > On Tue, 26 Mar 2002 04:23, you wrote: > > > I think removing the conditional that you suggest will have other > > repercussions. I will look further at this, but it will probably be a > > few days. > > > > john mcnally > > > > Rodney Schneider wrote: > > > Hi Edmund and John! > > > > > > I have sent two emails to this list regarding this bug in Intake and > > > no-one has replied yet :( It exists in both the most recent Turbine 2.2 > > > code and the most recent CVS version of Fulcrum. > > > > > > I'll try to explain myself a bit better this time as maybe the bug report > > > I sent wasn't clear. See this message... > > > > > > http://www.mail-archive.com/[email protected]/msg06959.html > > > > > > > > > I have got a problem with the rules for checking field values in > > > > > > intake. Somehow I cannot check for empty fields. I define the rule > > > > > > as explained in the docs like this: > > > > > > > > > > > > .. > > > > > > <field name="PersonName" key="personname" type="String"> > > > > > > <rule name="minLength" value="1">Bitte geben Sie den Nachnamen > > > > > > an.</rule> </field> > > > > > > .. > > > > > > > > > > > > When I leave the field empty, intake does not report an error. > > > > > > (isAllValid() still returns true) > > > > > > > > > > not sure why this does not work. > > > > > > If a browser submits a form with an empty text field (with, say, the > > > parameter name="foo"), then the request parameters will come in in the > > > form "?foo=&bar=..." etc., and both javax.servlet.HttpServletRequest and > > > org.apache.turbine.util.ParameterParser will report that the parameter > > > "foo" is present with the value "" (the empty string). However, > > > org.apache.turbine.services.intake.model.Field will treat this field as > > > if it is not present at all: specifically, in the method validate() is > > > the following code: > > > > > > stringValue = pp.getString(getKey()); > > > if ( stringValue.length() == 0 ) > > > { > > > set_flag = false; > > > } > > > > > > Thus, there is no way to distinguish between an empty field and a field > > > that is not present in the HTML form at all. > > > > > > I am happy to generate a patch that removes the above check, but I am not > > > sure if this will affect the rest of the Intake code... > > > > > > > > > I have experimented a bit with intake, and other rules actually do > > > > > > work as expected, if the field value is not empty. It seems the > > > > > > rules simply bypassed for empty fields and no checks are applied. > > > > > > So I also tried setting the required messsage: > > > > > > > > > > > > .. > > > > > > <field name="PersonName" key="personname" type="String"> > > > > > > <required-message>Bitte geben Sie den Nachnamen > > > > > > an.</required-message> </field> > > > > > > .. > > > > > > > > > > > > Again, no error when I leave the field empty. Is this a bug in the > > > > > > intake service (included in TDK 2.1 release) or am I doing > > > > > > something wrong? > > > > > > See above. > > > > > > > > <required-message> is to provide a message that is to be displayed if > > > > > the field is declared as required and the field is empty. It does > > > > > not make the field required. > > > > > > > > > > <rule name="required" value="true">Bitte geben Sie den Nachnamen > > > > > an.</rule> > > > > > > > > > > should work, though i have not looked at what's in 2.1 for a while. > > > > > > > > Thanks for the quick reply. That rule actually does seem to work > > > > (though I cannot find anything about a rule with name "required" in the > > > > intake docs at > > > > http://jakarta.apache.org/turbine/turbine-2/services/intake-service.htm > > > >l). > > > > > > The "required" rule does work, but this doesn't help in all situations. > > > One simple example is an "Edit User Profile" form which contains an > > > "Email Address" field that is not required and a "Work Phone Number" > > > field that is required. When a user first registers, they type in a > > > valid email address in the "Email Address" field and their current work > > > phone number in the "Work Phone Number" field. Then they change jobs, so > > > they go to the Edit User Profile page, empty the email address field, > > > edit the work phone number field (but accidentally type a 'abc' instead > > > of a valid number) and submit the form. Intake finds the the Work Phone > > > Number field is invalid so it redisplays the page containing the form > > > with the appropriate error message. HERE IS THE PROBLEM.... the email > > > address field is pre-populated with the old email address even though the > > > user had previously emptied it!! > > > > > > This happens because Intake doesn't disinguish empty fields from fields > > > that are not present in the form at all. See my other email for more > > > information. > > > > > > > A bit odd is, that the isAllValid() method still returns true, but > > > > lateron when I try to assign the fields to the proper data object > > > > (using the group.setProperties() method) an exception is thrown > > > > (message: "Attempted to assign an invalid input"). > > > > > > > > Maybe I can figure out another way to check the field validity (either > > > > that, or I will have to watch for that exception). > > > > > > I think the abovementioned code should be removed from > > > org.apache.turbine.services.intake.model.Field and > > > org.apache.fulcrum.intake.model.Field but, again, I am not sure whether > > > other code in Intake relies on the above check or whether existing > > > applications that use Intake will be affected. > > > > > > If you want me to submit a patch, please let me know. > > > > > > Thanks, > > > > > > -- Rodney > > -- > To unsubscribe, e-mail: <mailto:[EMAIL PROTECTED]> > For additional commands, e-mail: <mailto:[EMAIL PROTECTED]> -- To unsubscribe, e-mail: <mailto:[EMAIL PROTECTED]> For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>
