如何在 BizTalk 中处理 EDI 文件后执行一次存储过程?

发布于 2024-12-23 03:52:42 字数 213 浏览 2 评论 0原文

这更多的是一个一般的架构问题;我有一些想法,但我认为我们不能是唯一需要这样做的人。

当通过 BizTalk 处理 HIPAA EDI 文件(例如 837 医疗保健索赔)时,它会被拆分为多个索赔文件,并且它所经历的任何编排都会同时运行(嗯,实际上)。我需要在每个声明存储到数据库后只执行一次存储过程,因此该过程显然无法从保存数据的同一编排中执行。

建议?谢谢!

账单

This is more of a general architecture question; I have a few ideas but I figure we can't be the only ones who have ever needed to do this.

When processing a HIPAA EDI file through BizTalk (say an 837 healthcare claim), it is split in multiple claim files, and any orchestrations it goes through are run simultaneously (well, practically). I need to execute a stored procedure just once after every claim has been stored to the database, so the proc obviously can't execute from the same orchestration that saves the data.

Suggestions? Thanks!

Bill

如果你对这篇内容有疑问,欢迎到本站社区发帖提问 参与讨论,获取更多帮助,或者扫码二维码加入 Web 技术交流群。

扫码二维码加入Web技术交流群

发布评论

需要 登录 才能够评论, 你可以免费 注册 一个本站的账号。

评论(3

梦萦几度 2024-12-30 03:52:42

我可以想到一些选项,但我不知道如何保证您处理所有文档,而不按顺序处理文档或先检查文档以了解索赔数量。有几个选项:

  1. 您可以在编排中进行拆分并按顺序而不是并行处理它们,而不是让管道进行拆分。
  2. 再次使用单个编排进行拆分,但在开始拆分之前计算文档数量,并有一个单独的进程轮询要提交到数据库的文档数量。这将允许您至少异步触发所有子文档。
  3. 创建一个管道组件,在允许 EDI 组件进行拆分之前检查文档,并且与上面类似,对所有索赔进行单独的流程轮询以完成。

似乎这里最大的问题是,如果其中一项声明失败,您该怎么办,如果您只想在完整文件完成时提交,则需要一些回滚策略。

I can think of a few options but I don't know how to guarantee you process all of the documents without either processing the document sequentially or inspecting it first to know the number of claims. A few options:

  1. Rather than letting the pipeline do the splitting you could do the splitting in an orchestration and process them sequentially instead of in parallel.
  2. Again have a single orchestration do the splitting but before beginning the splitting count the number of documents and have a separate process that is polling for that number of documents to be committed to the database. This would allow you to at least fire off all of the sub documents asynchronously.
  3. Create a pipeline component that inspects the document before allowing the EDI component to do the splitting and similar to above have a separate process polling for all the claims to complete.

Seems that the biggest issue here is what do you do if one of the claims fails, you'll need some roll back strategy if you only want to commit when the full file completes.

回忆凄美了谁 2024-12-30 03:52:42

实现此目的的一种方法是在存储索赔的数据表上使用插入触发器。

一方面,这是满足您要求的非常简单的解决方案。但是,您必须确保其有据可查。触发器中内置的功能可能很难维护。

One way to do this would be by using an insert trigger on the data table where the claim is stored.

On the one hand this is a very simple solution to your requirement. However, you must make sure that it is well documented. Functionality, built into triggers can be hard to maintain.

梦幻之岛 2024-12-30 03:52:42

我们决定使用单个(而不是多个)HIPAA 模式,因此索赔文件将转换为包含多个 2000A 循环的单个 XML 消息。然后我们迭代这些循环,将它们映射进去,并在完成时触发过程。

We decided to use the single (rather than multiple) HIPAA schema, so the claim files is turned into a single XML message containing multiple 2000A loops. Then we iterate through those loops, map 'em in, and fire off the proc when complete.

~没有更多了~
我们使用 Cookies 和其他技术来定制您的体验包括您的登录状态等。通过阅读我们的 隐私政策 了解更多相关信息。 单击 接受 或继续使用网站,即表示您同意使用 Cookies 和您的相关数据。
原文