Set both "page" and "page_size" Params to PagedSearchIterable

[ashok2ashok]

Describe the bug
While trying to limit filename search results to a specific number - say 1, the library ends up iterating over all the pages until every result is collated. I was trying to limit my search result to only one page with a certain page_size. Not looking for the library to parse all possible pages.

To Reproduce
Steps to reproduce the behavior:

1. Perform a Search with a single filename, language and with a page size of 1. Sample given below:
   PagedSearchIterable<GHContent> searchResultsIter = github.searchContent().filename(fileName).language(language).list().withPageSize(1);

I followed the fix to this: #221 to come up with above code. There is no way to specify page param though.

2. Put a debug point on the below line (146 in v1.128) in GitHubPageIterator.java in fetch() method
   assert nextResponse.body() != null;
3. Verify the url param in the method at line 142
   URL url = nextRequest.url();
4. Resume Program to capture the looped page sizes. On every subsequent loop, the page is incremented automatically.

Expected behavior
To get the behavior of below GET URL through a way to specify page query param as well:
https://api.github.com/search/code?q=filename:ThreadSafeLazyLoadedIvoryTowerTest language:Java&per_page=1&page=1

Desktop (please complete the following information):

• OS: MacOS Big Sur 11.2.3, Mac mini (M1, 2020)
• Browser: Safari
• Version: 1.128

Additional context
I didn't find a way to limit my search results using both page_size and page query params. I tried to cheat by using the q param but Github doesn't not accept the page params in the query-field.
The first search API call doesn't not include page param.

The second auto-search API call added page param and increments it everytime until all results are fetched.

[ashok2ashok]

I now understand how silly my question was. I didn't notice the Iterable.

Fair warning to folks who are in the same boat - the issue will be masked and you'll be waiting forever if you use similar code to mine as shown below to read the results. searchResultsIter.forEach will continue calling the GH-API until the entire results are fetched instead of just the requested page_size. Same thing with searchResultsIter.toList()

           PagedSearchIterable<GHContent> searchResultsIter = github.searchContent().filename(fileName).language(language).list().withPageSize(1);
            List<GHContent> searchResultsList = new ArrayList<>();
            searchResultsIter.forEach(searchResultsList::add)

Since each call will exactly fetch the requested page_size, the iterator can be used to fetch the exact number of results you need - hopefully a multiple of page_size. To request records individually - use GHContent ghContent = ghContentPagedIterator.next() and to request entire page with the requested page_size use List<GHContent> ghContentList = ghContentPagedIterator.nextPage();

           PagedSearchIterable<GHContent> searchResultsIter = github.searchContent().filename(fileName).language(language).list().withPageSize(1);
            PagedIterator<GHContent> ghContentPagedIterator = searchResultsIter.iterator();
            if(ghContentPagedIterator.hasNext()){
                List<GHContent> ghContentList = ghContentPagedIterator.nextPage();
                //... Do you thing....
            }

Please note that PagedSearchIterable.getTotalCount() will make an additional call to the configured search - so beware :)

@kohsuke - Please close this issue as invalid or probably add this to the documentation :)
And can't thank you enough for this.

[kohsuke]

Iterable.forEach() naturally walks over the entire iterable and my feeling is that nobody should expect otherwise.

Thanks for a very thorough bug report, though.