Context
Part of the administrator side of our application requires the user to edit various types of content, which involves using a rich text editor or using files to generate content that can be seen by the 'client side' users of the application. It's kind of a domain-specific CMS lite.
Because this 'content' can be used in various parts of the application, it is included as a seperate relation in some of our domain entities. We decided to make our own tag library that defines some form fields that can be used to edit this content when an administrator edits an entity that includes a piece of content.
Question
What we'd like to be able to do is the following.
<form:form modelAttribute=...>
<olo:content-editor path="content"/>
<!-- Other form fields for this entity -->
<form:.../>
<form:.../>
</form:form>
The olo:content-editor tag then generates a number of form fields based on what type of content is needed. This means it may (or, depending on the type of content, may not!) generate the filebased-content tag which contains:
<input type="file" name="file/>
Which can be used to replace the file associated with the file based content.
The problem is that the Spring docs indicate that the file upload requires the form to have the enctype to define that it's sending multipart form data. As the file upload is part of the tag and not the form itself, we find this is undesirable. We would like to be able to use our olo:content-editor tag in forms without having to change the form enctype attribute. Is this possible?
Possible solutions
We can think of two client-side hacks that may resolve our problem, but both seem to be rather ugly solutions:
Include a script in filebased content tag that changes the form enctype when it's loaded, so that it is always set to the appropriate type. (Very ugly.)
Submit the file data as a regular hidden form field, of which the data is set by using the HTML5 File API (administrators use a compliant browser. This seems far less ugly but still not an optimal solution.)
Related
In a VUE.JS 2 application I'm working on, there is a series of form fields dynamically generated from a JSON Schema. As the Schemas change so do the inputs. (We are using a modified version of this library : https://gitlab.com/formschema/native/tree/feature/new-architecture.
I'd like to attach a Basic WYSIWYG to a particular textarea that was dynamically generated. All the libraries I've come accross need you to manually add the Tag to the template with attributes. Since the form is dynamically generated this seems difficult to do.
Is there a simpler vuejs way to attach a WYSIWYG Editor to a textarea? Perhaps by linking it to a certain ID. The textarea comes prepopulated with data from an API, and that info would need to be included. Any insight is appreciated. I've looked at the following libraries with no success:
https://github.com/ckeditor/ckeditor5-vue
https://github.com/froala/vue-froala-wysiwyg
https://github.com/davidroyer/vue2-editor
we are moving from content editor to experience/page editor as our primary editor interface and one of the fields we deal with is a raw multiline field for dropping custom script into the pages by editors. Eg campaign management, google tag manager or similar
Currently if I render a SC:fieldrenderer via webforms the script tags are becoming part of the dom on render.
What is the best approach to tackle this type of interface requirement via experience editor in SC7 webforms - it looks like sitecore isnt outputting to an input tag in order to supress and contain the code and is instead outputting to a span which is then flowing into the dom.
Thanks for any direction
I have tried to use Apache Tika v1.14 to parse text contained in PDF XFA forms. However, after trying different configuration in PDFParserConfig, I can only get the field names but not the text content of those fields.
For example, if there's a field called "Telephone", part of Tika's output may be <li fieldName="Telephone">Telephone: </li> (the same repeats for other fields as well). However, if I use pdfbox API to traverse the DOM tree to reach the Node named "Telephone", then I can use getNodeValue() to obtain the text content I want.
I am aware of the settings setExtractAcroFormContent() and setIfXFAExtractOnlyXFA() in PDFParserConfig and also experimented with them but I still didn't get the text content.
So my questions are
Did I misconfigure Tika so that it does not give the right output? Or,
Is this what Apache Tika's implementation is intended to do? Or,
Is the implementation still under development?
I am sorry that the forms contain medical information of patients so I am not able to attach them as example.
Thank you very much.
p.s. I am also aware of Tika's Jira issues https://issues.apache.org/jira/browse/TIKA-973 and https://issues.apache.org/jira/browse/TIKA-1857 so I thought this feature has been implemented.
I'm building a website with very small amounts of Javascript, just to add things to the page (like a contact form) without having to go to a new page.
I understand that I should build the contact page anyways (just in case the user doesn't have javascript turned on) and then use javascript if they've got it.
So where do I store the HTML for the form if I don't want to have it in two places?
(Normally I'm not so picky, but I'm curious on this one.)
If you have access to a server-side language, you can keep a separate snippet of the form in an external page. Then, you can include the snippet into the HTML content page with an appropriate include call. This has the added benefit that, for your JavaScript, you can pull the contact form from this snippet file using AJAX. In fact, many plugins allow you to display DHTML windows with HTML content. For example, check out ThickBox.
Without a server-side language, you can do something similar with frames. Just display the form snippet in a frame when you need to reference it. Personally, I don't like frames very much, so this isn't a very attractive solution for me, but you can use it if you choose (and style the frames appropriately).
Just put your HTML for the contact form in a .html file. Assuming you're using PHP or something, just include the file in your contact page and include it in the section for your dynamic contact form. The form should still submit to the same server-side page and have the same look and feel..
e.g. contactForm.html
<div class="contact-form">
<input ....>
</div>
I need to find a way to write a program (in any language) that will connect to a website and read dynamically generated data from the website.
Note that it's dynamically generated--it's not enough to get the source html, because the data I'm interested in is generated via javascript that references back-end code. So when i view the webpage source, I can't see the data. (For example, go to google, and do a search. Check the source code on the search results page. Very little of the data your browser is displaying is reflected in the source--most of it is dynamically generated. I need some way to access this data.)
Pick a language and environment that includes an HTML renderer (e.g. .NET and the WebBrowser control). Use the HTML renderer to get the URL and produce an HTML DOM in memory (making sure that scripting is enabled). Read the contents of the HTML DOM after the renderer has done its work.
Example (you'll need to do this inside a System.Windows.Form derived class):
WebBrowser browser = new WebBrowser();
browser.Navigate("http://www.google.com");
HtmlDocument document = browser.Document;
// extract what you want from the document
I used to have a Perl program to access Mapguide.com to get the drive direction from one location to another location. I parsed the returned page and save to database. If the source never change their format, it is OK. the problem is the source format often change, your parser also need change.
A simple thought: if we're talking about AJAX, you can rather look up the urls for the dynamic data. Then you can use the javascript on the page you're talking about to reformat this.
If you have Firefox/greasemonkey making a DOM dumper should be a simple matter.