How can I solve the problem of unfeasible add-ons in SVN ?, Svn cannot add-on

How can I solve the problem of unfeasible add-ons in SVN ?, Svn cannot add-on

There is a saying that "the effect is good, who knows who to use". During the cooperation between the data center and SVN, the team lead was responsible for the project submission and did not participate in the project. I only knew the updates, I don't know about compilation and submission, but when I recently implemented the score system, I was responsible for one of the three solutions because our project was divided into three solutions, after the project test is successful, the framework is basically complete. I thought it would be enough to directly upload the project to SVN, but what I couldn't think of was the problem that followed.

The purpose of writing is to identify by yourself and avoid mistakes.

1. Problems Encountered

The upload can be successful, but I will be dizzy when I check out it. What is the problem? See:

Thank you for your help. Let's take a look at it.

The first attempt to solve the problem:

I thought that the release was not successful, so I made numerous releases and found that it could not be loaded, but other layers were acceptable, such as the client and the WCF layer. When we remove the so-called non-add-ons and load them again, we find that there is no problem, however, it is still not feasible to regenerate the solution or run the previously written program. So the first attempt failed.

Ø second attempt to solve the problem:

When we open the non-loaded attribute items (right-click), we find that the loading path is not the path released to SVN, but the previous path.

Because our system is to build three solutions, layer B and layer D are public, so for layer B and layer D, I use to load existing items, then copy the corresponding B and D layers. The result is that the program I uploaded to SVN loads the upload path, however, the upload path cannot be used to load the existing items of layer B and layer D. As a result, the path cannot be identified or the corresponding path cannot be loaded, the path is found when I remove it and add it again, so it can be loaded.

Therefore, exercise caution when copying data. If you only use the data locally, you only need to reference the data. However, if you upload the data, you must copy the data and add it again, this ensures that the path is correct, and the path is still correct even when uploaded to the server.

Ø solution to the third attempt:

After all the previous work was done, I thought it would be safe to upload again, but the error still occurred. The error prompt is obvious, that is, the address of the referenced library file does not exist, at this time, I suddenly realized that my reference file is also an existing item added. If it is local, it is possible to find the file path on the computer, in this way, the corresponding library file can be generated under the corresponding binfile, so that the program can run successfully, but if it is not local, the path for adding existing items cannot be found on SVN. Therefore, an error occurs, and the library files under the binfiles in each layer cannot be generated. As a result, the program runs or cannot be generated, so we 'd better copy all the external files needed to the package. Then, the corresponding files are searched from the package to ensure security and availability.

2. Summary

In a word, copying is very easy. Remember to be cautious. When you encounter problems and solve the problems, we have been growing.