/// <summary>
/// Parses a batch of jobs
/// </summary>
/// <param name="apiClient">The API client to use to parse the files</param>
/// <param name="parseOptions">Any parsing/indexing options</param>
/// <param name="rules">
/// The rules that should be applied to whatever files are found prior to parsing.
/// This is important to reduce the number of invalid parse API calls and reduce parsing costs.
/// </param>
/// <param name="directory">The directory containing the files to be parsed</param>
/// <param name="searchOption"></param>
/// <param name="successCallback">A callback for when a file is parsed successfully</param>
/// <param name="partialSuccessCallback">A callback for when some error happened during/after parsing, but there is still usable data in the response</param>
/// <param name="errorCallback">A callback for when an error occurred when parsing the file, and there is no usable data</param>
/// <param name="generateDocumentIdFn">A callback so you can specify a DocumentId for each file that is parsed</param>
/// <exception cref="SovrenInvalidBatchException">Thrown when the directory provided does not meet the <see cref="BatchParsingRules"/></exception>
public static async Task ParseJobs(
SovrenClient apiClient,
ParseOptions parseOptions,
BatchParsingRules rules,
string directory,
SearchOption searchOption,
Func <JobBatchSuccessResult, Task> successCallback,
Func <JobBatchPartialSuccessResult, Task> partialSuccessCallback,
Func <BatchErrorResult, Task> errorCallback,
Func <string, string> generateDocumentIdFn = null)
{
if (apiClient == null)
{
throw new ArgumentNullException(nameof(apiClient));
}
IEnumerable <string> files = GetFiles(rules, directory, searchOption);
//process the batch serially, since multi-threading could cause the customer to violate the AUP accidentally
foreach (string file in files)
{
Document doc = new Document(file);
string docId = generateDocumentIdFn == null?Guid.NewGuid().ToString() : generateDocumentIdFn(file);
try
{
//set document id if we plan to index these documents
if (parseOptions?.IndexingOptions != null)
{
parseOptions.IndexingOptions.DocumentId = docId;
}
ParseRequest request = new ParseRequest(doc, parseOptions);
ParseJobResponse response = await apiClient.ParseJob(request);
if (successCallback != null)
{
await successCallback(new JobBatchSuccessResult(file, docId, response));
}
}
catch (SovrenUsableJobException e)
{
//this happens when something wasn't 100% successful, but there still might be usable data
if (partialSuccessCallback != null)
{
await partialSuccessCallback(new JobBatchPartialSuccessResult(file, docId, e));
}
}
catch (SovrenException e)
{
if (errorCallback != null)
{
//this happens where there is no usable data
await errorCallback(new BatchErrorResult(file, docId, e));
}
}
}
}
internal JobBatchSuccessResult(string file, string docId, ParseJobResponse response)
: base(file, docId)
{
Response = response;
}
internal SovrenProfessionNormalizationJobException(RestResponse response, ApiResponseInfoLite errorInfo, string transactionId, ParseJobResponse parseResponse)
: base(response, errorInfo, transactionId, parseResponse)
{
}
internal SovrenIndexJobException(RestResponse response, ApiResponseInfoLite errorInfo, string transactionId, ParseJobResponse parseResponse)
: base(response, errorInfo, transactionId, parseResponse)
{
}
internal SovrenUsableJobException(RestResponse response, ApiResponseInfoLite errorInfo, string transactionId, ParseJobResponse parseResponse)
: base(null, response, errorInfo, transactionId)
{
Response = parseResponse;
}
