Relative URLs and Application Context

How do you handle rendering relative URLs so that:
1. Your application is independent of the application context it's deployed to
2. You don't end up with lots of ugly code in your views
3. The solution will work on any J2EE server
I have several approaches, but I can't find any "best practices" discussions on this subject so I thought I'd start one.

<%@ taglib prefix="c" uri="http://java.sun.com/jstl/core"%>
"> Home</a>
My main problem with that is that it's ugly as sin.

Similar Messages

Can't view dashboard with relative url in BI 4 Launchpad

I created a dashboard using BIWS connections and I am able to preview the dashboard when i have the full path of Web Service URL but when I use the relative url and publish to launchpad I get the "Cannot access external data......, Unable to load URL:/dswsbobje/Services/Session..." error message. Any ideas on how to solve this?
Thanks

Are you trying to preview the dashboard from dashboard designer tool ? If yes, then you must have a full web service url, otherwise, it will not be able to resolve the server and port.
But, if you use relative URL and export the same to BO repository, you will be able to view the same by logging into BI Launchpad. There it can resolve the server name and port from existing session.

How to convert relative url to absolute in URLPortlet?

Images and links don't work becase they are not getting filtered to absolute urls. <baseHRef> doesn't seem to do it. Isn't URLServices supposed to automatically convert all urls to absolute?

URL Services internal identifies relative URL and converts them into absolute URL's.
It uses <BASE HREF="URL"> if available, otherwise the uses <pageUrl> as base URL and constructs the absolute URL's.
Please post URL to which you facing problem, so that we can find fix this error.

Relative urls

I am writing a program and want to use a relative url in case the other parts of the address change in the future. Right now I have this code in my applet:
getAppletContext().showDocument(
new URL("adam.ThePopup"), "popupWindow");But nothing happens. Is there a way to imply that you want the relative url and not specify the whole address? Thank you.

use getDocumentBase()
as per API of class Applet:
public URL getDocumentBase()Returns an absolute URL naming the directory of the document in which the applet is embedded. For example, suppose an applet is contained within the document:
http://java.sun.com/products/jdk/1.2/index.html
The document base is:
http://java.sun.com/products/jdk/1.2/

ScriptBundles and application relative URLs

I'm having trouble coming up with a way to bundle javascript files in my environment using .net's scriptbundles. My environment is setup with one website and about 30 applications configured in IIS under this one site. I also have about 20 or
so different custom javascript libraries that provide various features or "widgets" to the applications. The widget's .js files are located in folders at the root of the site. Each application may use some or all of these widgets.
While I have gotten the bundles to work for .js files that are local to each applcaition, I'd like to bundle together the .js files that are external to the application. This would allow me take advantage of the minifying and help resolve any caching
issues.
Is there any way possible to not have to use application relative URLs?
I have looked at trying to use the CDN option, but that doesn't work, because while in debug mode it tries to use a local version of the widget and there will never be a version local to the app, but there will always be one at the site root.
I also noticed that you can override the .Include function of the ScriptBundle class. Maybe there's an answer there, but I couldn't figure it out.
It seems to me that constraining the files that can be included to just the files located in the application is poor judgment.

Hello TroySim,
It seems this issue is related with a web application, I suggest that you could post it to:
http://forums.asp.net/
The current forum you posted to is used to Discuss and ask questions about .NET Framework Base Classes (BCL) such as Collections, I/O, Regigistry, Globalization, Reflection.
And anyway, it is glad to hear that you find a workaround for this issue.
Regards.
We are trying to better understand customer views on social support experience, so your participation in this interview project would be greatly appreciated if you have time. Thanks for helping make community forums a great place.
Click
HERE to participate the survey.

How to set up use of relative URLs for a BSP application

Dear all,
I need to access a BSP application through our external portal.
This is failing because generated URLs are absolute URLs (they mention physical server name, not known of course on the internet) where they should be relative URLs (they use external portal URL as prefix).
How to set this up?
I'm logged on http://myportal.com/irj/portal and want to call a BSP application. To build this page, my internet client (Internet Explorer) tries to get elements from the server. Looking at HTTP trace, I can see that my IE is trying to reach the physical server, not known on internet of course, then page loading fails.
Ex absolute URL : http://myservername.domain.com/sap/bw/[...]/image1.gif
This is the behavior I can see at the moment.
Relative URL : http://myportal.com/sap/bw/[...]/image1.gif
This is what I'd like to see.
Many thanks in advance for your help.
François

Any idea ?

Relation between URL and content in JSF navigation

Hello,
I've tried lots of JSF examples and there always occoured a strange kind of navigation to me:
The URL you see in the browser's address bar has nothing to do with the actually displayed JSP. Instead of that, the URL is always pointing to the previously called JSP - where the call to the current JSP comes from.
I know where this comes from:
If the form tags don't define an action, they are pointing to the current page itself.
And the actually called navigation case is chosen by a (hidden) post parameter.
So the JSF controller servlet doesn't actually care, which URL is called - it just regards the mentioned parameter.
This kind of navigation seems quite confusing, especcially since JSF claims to make navigation clearer.
One of the basic demands of the web applications we are developing is, to have a clear relation between URL and page content.
E.g. it must be possible to index the pages with search engines. Or to simply bookmark distinct pages in the browsers favourites.
Another problem about this navigation concept is, that it always works with forms and with post method (you can't choose get method), even if you just have simple links - e.g. detail view of a single item after selecting it from a list of items. That way it's not possible to ever use the back button of your browser.
I'm new to JSF, so my question:
Is this the (one-and-only) navigation concept of JSF, or are there alternative concepts as well?
I would be surprised, since the mentioned demands are basic web standard - although I didn't find any alternatives so far.
Thanks,
Chily

Thanks for the answer!
I'm not sure, if using a redirect on every single call is very achieveable and in the purpose of the JSF design.
E.g. it is generating a lot of traffic overhead.
(Identifying the navigation case by the page name instead of a hidden parameter would help a lot.)
Or the history of the browser doesn't work with it, so you can't identify visited links.
But anyways, there are also internal problems with it in most cases: I've inserted the <redirect/> tag in the navigation cases of a few examples. This just causes a call of the <to-view-id> page, without any parameters. So I probably have to add some non-standard-processing around the call to make it work:
If e.g. I have the standard case of a list of items (let's say in a dataTable), which are identified by an id. And each item uses a commandLink to call it's detail view.
* How do I get the identifier (id) as a parameter into the redirect URL?
* And how do I access this parameter in the destination form bean to choose the item to be displayed? (see also http://forum.java.sun.com/thread.jspa?threadID=607505)
Thanks for help,
Chily

How to define JAXP Parsers and Transformer in a Web Application Context

Hi
I need to define JAXP Parsers and Transformer in a Web Application Context where I can deploy this application in any J2EE Application Server without modifying the server. That means that I can't set System properties or set the jaxp.properties file. I tried to use the META-INF/services/javax.xml.transform.TransformerFactory file but it didn't work.
Any ideas?

Hi
I need to define JAXP Parsers and Transformer in a Web Application Context where I can deploy this application in any J2EE Application Server without modifying the server. That means that I can't set System properties or set the jaxp.properties file. I tried to use the META-INF/services/javax.xml.transform.TransformerFactory file but it didn't work.
Any ideas?

Correctly combining an absolute and a relative URL

Hello everyone!
I've been messing around with combining absolute and relative URLS in java and have found the task to be a little more confusing than I originally thought. For example:
/wallpapers/17i/Lotus_Exige_Mk_II,_2004.jpg
http://wallpapers.diq.ru/ru/17__Peugeot_Moonster_Concept%2C_2001.html
Firefox and other browsers see this as: http://wallpapers.diq.ru/wallpapers/17i/Lotus_Exige_Mk_II,_2004.jpg
I don't understand why the /ru/ is taken out; and was wondering if there was already a class written in Java that combined urls, or if someone could explain to me why the end result is not http://wallpapers.diq.ru/ru/wallpapers/17i/Lotus_Exige_Mk_II,_2004.jpg
Thanks

The web page could have a <head> section that contains a [<base>|http://www.w3schools.com/tags/tag_base.asp].
This specifies the base that will be used when resolving urls. It may not be the url of the page itself. For instance it could be that the "ru" bit collects together all the Russian langauge pages parallel to a "en" duplicate that has the same pages in English. Clearly both sets would want to access the same image resources so they have a <base> setting that points one level higher up.
If this is the case and you want to get the url for the images you will have to parse the <head> of the web page and figure out what the <base>is.

TMG / WPAD and protocol relative URLs

We are using a TMG Cluster (TMG2010 SP3) as a proxy.
We use wpad auto discover for our clients and that is working fine except that protocol relative urls are not sent to the proxy but use the direct route.
More and more websites are using a protocol-relative url (eg: script src=//ajax.googleapis.com/ajax/libs/jquery/1.9.1/jquery.min.js ) when we use the "automatically detect settings" option in IE these URL's aren't redirected to the proxyservers.
And because our clients do not have direct internet access the page doesn't load the script and isn't displayed correctly.
I probably have to alter the wpad.dat file but I'm stuck in how to do so.

Hi,
Thank you for your post here.
How did you judge that the packets have been directly routed to gateway?
Could you please configure your proxy to manually point your web proxy server address? Please test it to see if traffic is still routed directly.
Network Monitor 3.4 is a very useful tool to help us capture the network packets. We need to check if the request you send is packaged in HTTP or HTTPs since just HTTP or HTTPs can be proxy direct to proxy server.
Best Regards
Quan Gu

SharePoint sites and the use of Relative URLs

In our sites the designers coded all URLs as FQDN (e.g.
http://servername/sites/mysite1....). This includes URLs in webparts too. We now want to transfer the site to a new SharePoint box, of a different name, and power down the original box. We ran a test and, sure enough, the FQDN URLs
hardcoded in the copied webparts point back to the original box. This means when the old box is eventually powered down the new one won't work. My question is this: should our Designer have used relative URLs (e.g. ../sites/mysite1) everywhere
instead of referencing the server name?
TIA,
edm2

yes... absolutely.
Better yet, there are special URLs that designers can use, which are relative to the SITE COLLECTION, or even the SITE/SUBSITE... which are really better to use, should you ever need to copy the site collection, or change its URL.
Also, SharePoint should use an ALIAS name, not a server name.
Scott Brickey
MCTS, MCPD, MCITP
www.sbrickey.com
Strategic Data Systems - for all your SharePoint needs

Difference between application context and servlet context

HI,
what are the differences between an application context and a servlet context?
Also can any body elaborate on context path and servlet path. To quote servlet spec:
Context path: The path prefix associated with the ServletContext that this
servlet is a part of.
Servlet Path: The path section that directly corresponds to the mapping
which activated this request.Thanks in advance.

Application context and Servlet Context are pretty much interchangeable terms IIRC.
ContextPath == name of your web application
ServletPath == path to the servlet
Given http://localhost:8080/myWebApp/myServlet
contextPath == myWebApp
servletPath == myServlet
A jsp to help you understand this:
Working with server: <%= application.getServerInfo() %> 
Servlet Specification: <%= application.getMajorVersion() %>.<%= application.getMinorVersion() %> 
Sessionid : <%= session.getId() %> 
QueryString: <%= request.getQueryString() %> 
ContextPath: <%= request.getContextPath() %> 
ServletPath: <%= request.getServletPath() %> Cheers,
evnafets

Lync reporting: Found QOEMatrics database stores media related info of all interactions(Audio,Video ,conferences) except IM and application sharing.

Do we have any way to let QOEMatrics db to store media info of every sessions(including IM,sharing)?
Ref:Lync 2010 SQL reporting databases and tables.

The following types of content are not archived with Archiving Server:
Peer-to-peer file transfers.
Audio/video for peer-to-peer instant messages and conferences.
Desktop and application sharing for peer-to-peer instant messages and conferences.
For the detailed information, see
http://technet.microsoft.com/en-us/library/jj204900.aspx
Lisa Zheng
TechNet Community Support

MVC model in Web systems and applications

MVC model in Web systems and applications
Object-oriented design model is experience, MVC idea is a user interface for the original. This article discusses how the major application areas in the new Web design patterns and the use of MVC framework. The article first introduced the concept of design patterns and characteristics, and MVC architecture design concepts and analysis of the MVC framework contains several key models. Based on the characteristics of Web applications on how to use patterns and MVC framework made some design ideas.??
1. Introduction
1.1 design model
Object-oriented technology and the emergence of software applications has greatly enhanced the trusted and software quality. Object-oriented programming than previous models to various programming simple and efficient, but the object-oriented design methodology than the previous design methods to complex and much more skill, a good design should be both on the issue of gender, but also to take full account of the problems and needs sufficient interoperability. In the past 10 years, people in the object-oriented technology and the practical application of research to explore certain issues in relation to the creation of a number of good solutions, the so-called object-oriented design patterns. Object-oriented technology is one of the purposes of enhancing the software trusted, and to design model, design programmes in important positions from a deeper sense of meaning and essence embodies trusted. There are many people in the design model definition, which cited Christopher Alexander is the largest design model definition : Each design model is a tripartite rule, which expresses a contextual environment (Context), a problem and a solution. Design models generally following basic elements : model name, the purpose solution effect 1995-1998 code and related design models. There are several classifications design patterns can be divided into a model based on the purpose (Creational), structural type (Structural) and the type of behaviour (Behavioral) three. It is mainly used in the creation of a model-based object model-based structure to deal primarily with the category or combination of objects, used to describe behavior-based model is the main target for the category or how stress and how to allocate responsibilities. Design patterns can be divided into categories based on the scope and target mode model type model dealing with the relationship between the categories and sub-categories, these relations through the establishment of succession in Translation moment to be finalized, are static. Model is targeted at addressing the relationship between the moment of change these relations in the operation, more dynamic. Model features : through the experience acquired in a structured format to write down, avoid encountering the same problems on the first design, exist in different abstract level, in continuous improvement, can be trusted artificial product for the design and best practice in the world to be combined to address larger issues.
1.2 MVC framework
MVC was first used in a user interface Smalltalk-80 China. M representative models Model, representatives maps View V, C representatives controller Controller. MVC trusted code with the aim of increasing the rate of data reduction expressed, the data describing the operation and application coupled degrees. Also makes software Keweihuxing, restorative, expansionary, flexibility and packaging of greatly enhanced. Single-user applications are usually incident-driven user interface to the organizational structure. Staff development tool with an interface painting of a user interface interface code based on user input and then prepare to implement the corresponding moves, many interactive development environment encouraged to do so, because it emphasizes first and then a functional interface. Some software design model is the strategy that will be fixed before the code into the regular system of the final. Result is that the procedures and organizations around the user interface elements in the user interface elements of those moves, data storage, applications and functions of the code is used to indicate the way intertwined. In single-user system code structure can be so, because the system will not demand frequent changes. But for a large system such as large Web systems, or e-commerce systems to be applied. Model by incorporating data from a variety of access and control data can be separated to improve distributed system design. MVC design pattern is composed of three parts. Model is the application object, no user interface. Type in the screen showing that it represents the flow of data users. Controller user interface definition response to user input, the users responsible for the action against the Model into operation. Model View data updated to reflect the adoption of data changes.
2. MVC design pattern,
An MVC framework for the design of the system includes many models, but with MVC is most closely related to the following three models : Observer, Cambridge and Strategy.
2.1 Observer models
MVC through the use of purchase / notification form and the separation of the Model View. View to ensure that their content accurately reflected Model and state. Once Model content changes, there must be a mechanism to allow notification to the relevant Model View, View can be made relevant at the appropriate time updating of data. This design is also more general problems can be solved, the target separation, making a change to the target audience affect others, which targets those who do not know the details of the object being affected. This is described as Observer in the design model. Model type : Observer model is the object-oriented model, it is behaviour-based model. Model purposes : definition of hierarchical dependence relations between objects, or when a target value of the state change, all its dependent relationship with the object are notified and automatically updated. There are a variety of data may show a way, in different ways and may also show. When a way through a changed data, then the other should be able to show immediately that the data change and do accordingly.
Effect :
1. Abstract coupling. I only know that it has a target audience of some observers, the observers met each abstract Observer category simple interface, does not know their specific affiliation categories. This makes the coupling between goals and observers smallest and abstract.
2. Support radio communications. Needless to notify designated observers goals, how to deal with the observer informed decisions.
3. Possible accidents updated. We update logic, avoiding mistakes updated.
2.2 Faculty model
MVC is an important feature of View can nest. Nest can be used for any type of combination of local maps available, but also management of type nest. This thinking reflects the type and mix of components will be equal treatment design. This object-oriented design ideas in the area of Cambridge has been described as a design model. Model types : Cambridge model is the object-oriented model, it is also the structure type model. Model purpose : to target portfolio into tree structures to express "part-whole" level structure. Prepared for the use and combination of individual target audiences with the use of consistency.
Effect :
1. Definition of a target portfolio includes simple objects and the structure of the category level. Simple objects may be complex combinations of objects, and can be targeted portfolio mix. This customer-code used in the target areas can use simple combinations target.
2. Simplify customer-code. Needless to know their customers - a mix of target audiences is a simple target or can use these items in a consistent manner.
3. Easier to add new types of components. New components can easily be changed to a combination of customer-targeted codes.
2.3 Strategy model
Another important characteristic is the MVC can not change the View of changes View response to user input. This often requires a change in response to the logic of the system is very important. MVC to respond to the logic involved in the Controller. Controller of a category level structure could easily change to the original Controller appropriate, a new Controller. View Controller son used to achieve a specific example of such a response strategy. To achieve different response strategy, as long as examples of the use of different types of replacement will Controller. Also in the running time by changing the View Controller for users to change View of response strategies. This View-Controller relationship was described as an example of Strategy design pattern. Model types : Strategy model is the object-oriented model, it is behaviour-based model. Model purposes : definition of a series of algorithms, and their packaging, and ensure that they can replace each other, making algorithms can independently use its customer-change.
Effect :
1. Strategy category levels for Context definition of the relevant algorithms can be trusted or behaviour.
2. Alternative methods of succession. If the direct successor Context, with different acts will be added Context act, the realization of which would algorithm mixed up with Context, Context hard to preserve and expand, but can not dynamically changing algorithms. Will be enclosed in a separate Strategy category algorithms to enable algorithm independent Context change easily cut over expansion.
3. Can provide the same acts different date.
4. Strategy-must understand what customers between different.
5. Context and Strategy communications between costs.
6. An increase in the number of targets.
3. MVC in Web application system
Now some of the distributed systems such as Web-based B2B e-commerce system, suitable for use MVC framework. Through analysis from the perspective of high-level applications can be a target divided into three categories. Category is shown for the target audience consists of a group of commercial rules and data, there is a category that is receiving requests to control commercial target to complete the request. These applications often need to change is shown, such as web style, color, but also need to demonstrate the contents of the display. And the business rules and data to be relatively stable. Therefore, said that the frequent need to change the View objects that the business rules and data model to be relatively stable target, and that the control of the Controller is the most stable. When the system is usually issued after the View objects by artists, designers or HTML/JSP system managers to manage. Controller target applications development personnel from the development and implementation of rules for commercial and business development personnel from the target data, database managers and experts in the field of common completed. Show in Web?? or customers - control logic can be Servlet or JSP, dynamically generated Html. Generally used Servlet better than using JSP. JSP will be better with the Html code of separate codes for page designers and developers of separation efficiency. Servlet and JSP can complete all complete functions, actually JSP eventually converted into a Servlet. And control of the target system exists in every level, the coordination of cross-layer moves. Contain business rules and data objects exist in the EJB layer (EJB-centred model) or Web?? (Web-centred model).
3.1 View in the Web application system
View of the system shows that it fully exist in Web??. General by JSP, Java Bean and Custom Tag. JSP can generate dynamic web content using Java Custom Tag easier Bean, but it can show the logic of packaging, and more conducive to modular trusted. Some well-designed in a number of JSP Custom Tag can even be used in different system duplication. Model for control of JSP and Java Bean objects. JSP through Java Bean objects to retrieve the data model, the Model and Controller object is responsible for updating the data on Java Bean. In general, can we devise all possible screen that users can see all the elements of the system. Based on these elements, to identify the public part of passive components and dynamics. Can consider the use of templates means to separate the content generated JSP public, also need to change their generation Html or JSP from a JSP templates to dynamically introduce these different parts (include methods). Another issue to consider is screen option, when dealing with End users request template automatically available to show that the concern that users must know what is the screen components. So can consider all screens on the definition of a centralized document, such as a document or text document java. Taking into account the possibility of changes in future document definition screens, the best use of text documents such as a XML document, so future changes to the recompilation. According to the URL and user input parameters to shine upon the results of a screen, of course, likely to be made on the basis of the outcome of the implementation of actions to choose different results screen. Therefore, the need for a request for matching resources with document (XML), if a URL request several different results, it must specify in the document need to control the flow (a controller object), as well as the corresponding screen different flows.
3.2 Model in the Web application system
Model objects represent business rules and business data exist in EJB layer and Web??. In J2EE norms, the system needs some data stored in the database, such as user account information (account model), the company's data (company model), some not recorded in the database. If a user browsing the current catalogue (catalog model), the contents of his shopping (shopping cart model). Which one of these models exist in the data according to their life cycle and scope to decide. In Web?? a HttpSession and ServletContext and Java Bean objects to store data in the EJB layer is a data storage and logic EJB to. Web?? the Java Bean objects stored in the model layer model of the EJB object data copy. Because there are many different EJB tier model targets, so Web?? through a ModelManager to control the EJB layer object model in ModelManger background model can be used for packaging methods. In the EJB layer and the rules have all the data into EJB model is inappropriate. If the database can visit the Dao object model into objects. Dao can be encapsulated and the specific details of the database in the world, if we can write a different table, a number of databases, or even multiple databases. If the orders can be a model for OrderDAO, it may have to deal with Order table, table and OrderItemLines OrderStatus table. Value can also consider the use of targets. Value can be a target of securing long-range targets, because every time the remote object attributes could be a long-range redeployment process will consume network resources. EJB objects in the distance can be used instead target. In the distance, one-time items to be targeted instead of the value of all attributes.
3.3 Controller in Web application system
Coordination with the Model View Controller object to the request of users into the system to identify incidents. In Web?? generally a MainServlet (or Main.jsp), and receiving all requests, it can use screen flow management devices (ScreenFlowManger) decided next screen. There is a general request processors RequestProcessor contains all requests are needed to be done to deal with logic, such as the request translated into system events (RequestToEvent). Acting processors usually also includes a request for ClientControlWebImpl, it is logical to deal with the EJB layer in Web?? Acting. In EJB layer, a layer for EJB tier Web ClientController provide the CD visit. Another StateMachine used to create and delete ejb handle Web?? sent to the incident. Controller Another important function is synchronous View and Model data. ModelManger contained in a ModelUpdateManger, it puts events into a Model System assembly that all the needs of synchronous Model, and then notify Listeners do synchronous operation.
4. Concluding remarks
In recent years, with the Internet technology development and the emergence of new business models, the Web is based on a large number of applications. On how to design these systems architecture, and gradually there has been some convergence of opinion, the most important point is that its structure should be rational in the open. Demand than ever faster development of technology and design concepts, systems for the future, the cost of upgrading the smallest, research software systems architecture still very useful and necessary.

Bravo. And your point is?

Sharepoint 2013 Wiki Page - Links in content rewritten to relative urls

We have an issue where on all SharePoint 2013 wiki pages any URL to items within the site containing the wiki page are rewritten to be relative URLs. This causes a huge problem when we surface these pages and a snippet of their content through a search web
part because the links fail being relative to the wiki page itself. To further illustrated with an example.
A site collection root page http://www.sp2013.com/default.aspx contains a search results web part returning all wiki pages in the site collection. There is a custom display template that renders
the linked title of the wiki page and the first 500 characters of formatted text (html) of the wiki page.
A wiki page is created in a sub site (sub1) called wiki1 (http://www.sp2013.com/sub1/pages/wiki1.aspx) the text of the page contains:
"Check out our new library: Document Library"
Document Library is a hyperlink to (http://www.sp2013.com/sub1/Document%20Library/Forms/AllItems.aspx)
Upon saving the wiki page the link is reformatted to be /Document%20Library/Forms/AllItems.aspx
From the rollup web part at http://www.sp2013.com/default.aspx the relative url in the snippet text of the display template evaluates to
http://www.sp2013.com/Document%20Library/Forms/AllItems.aspx which does not exist.
I have attempted modifying the source of the wiki page as well as the link in the ribbon however upon save of the changes the link is always modified to be relative. Any assistance is appreciated as this is extremely frustrating. Thank you.

Hi JonesEJ,
According to your description, my understanding is that SharePoint converts a absolute URL to
a relative URL when you save the wiki pages containing the URL.
This is actually the preferred behavior in SharePoint and on the web in general. SharePoint saves all absolute URLs for the wiki content as relative. It only does this if the URL matches the URL for the web application for that current wiki
site.
There is no OOB way to change the relative URL back to the absolute URL. For your issue, you can try to use
SPSite.MakeFullUrl() to convert it back.
There is a similar post for your reference:
http://sharepoint.stackexchange.com/questions/62887/absolute-urls-converted-to-relative-urls-when-saving-item
Best Regards,
Wendy
Forum Support
Please remember to mark the replies as answers if they help and unmark them if they provide no help. If you have feedback for TechNet Subscriber Support, contact
[email protected]
Wendy Li
TechNet Community Support

Relative URLs and Application Context

Similar Messages

Maybe you are looking for