Incrementing and initializing multiple counters using Java Streams

Please I need help with the use of streams to initialize counters while looping through a collection:

//Method containing the logic to be converted to streams

    int errorOnlyCount = 0;
    int warningOnlyCount = 0;
    int bothErrorAndWarningCount = 0;
    int successfulProcessedCount = 0;


for(ReportItem summaryReportItem: summaryReportItemList){
    Map<String, Integer> summaryReportItemMap = getTermErrorCount(summaryReportItem);

    if(summaryReportItemMap.get("errorCount").intValue() > 0 &&
            summaryReportItemMap.get("warningCount").intValue() == 0
    ){
        errorOnlyCount += 1;
    }

    if(summaryReportItemMap.get("errorCount").intValue() == 0 &&
            summaryReportItemMap.get("warningCount").intValue() > 0
    ){
        warningOnlyCount += 1;
    }

    if(summaryReportItemMap.get("errorCount").intValue() > 0 &&
            summaryReportItemMap.get("warningCount").intValue() > 0
    ){
        bothErrorAndWarningCount += 1;
    }

    if(summaryReportItemMap.get("errorCount").intValue() == 0 &&
            summaryReportItemMap.get("warningCount").intValue() == 0
    ){
        successfulProcessedCount += 1;
    }
}


//Method called from the logic above

private Map<String, Integer> getTermErrorCount(ReportItem summaryReportItem){
        int errorCount = 0;
        int warningCount = 0;

        if(summaryReportItem.getIsDuplicateItemError())
            errorCount += 1;
 
        if(summaryReportItem.getIsDescriptionWarning())
            warningCount += 1;

        Map<String, Integer> errorWarningCounterMap = new HashMap<>();
        errorWarningCounterMap.put("errorCount", errorCount);
        errorWarningCounterMap.put("warningCount", warningCount);

        return errorWarningCounterMap;
    }

//POJO with the string id and booleans variables

public class ReportItem {
        private String id;
        private boolean isMissingIdError;

        public ReportItem id(String id) {
            this.id = id;
            return this;
        }

        @Nonnull
        @JsonProperty("id")
        @JsonInclude(JsonInclude.Include.USE_DEFAULTS)
        public String getId() {
            return this.id;
        }

        public void setId(String id) {
            this.id = id;
        }

        public ReportItem isMissingIdError(Boolean isMissingIdError) {
            this.isMissingIdError = isMissingIdError;
            return this;
        }

    @Nullable
    @JsonProperty("isMissingIdError")
    @JsonInclude(JsonInclude.Include.USE_DEFAULTS)
    public Boolean getIsMissingIdError() {
        return this.isMissingIdError;
    }

    public void setIsMissingIdError(String id) {
        this.isMissingIdError = isMissingIdError;
    }
}

I was thinking of creating a stream for each if-block, but then how do I increment and initialize the counters.

Below is my code for the first counter's block.

 Stream<ReportItem> errorOnlyCountStream = summaryReportItemList.stream().filter(i -> (getTermErrorCount(i).get("errorCount").intValue() > 0 &&
                getTermErrorCount(i).get("warningCount").intValue() == 0));

Please, can someone help out with a snippet for the best approach?

You can actually do this all reasonably well from one stream.

Map<Boolean, Map<Boolean, Long>> results =
   summaryReportItemList.stream()
     .map(ThisClass::getTermErrorCount)
     .collect(
       Collectors.partitioningBy(
         map -> map.get("errorCount") > 0,
         Collectors.partitioningBy(
           map -> map.get("warningCount") > 0,
           Collectors.counting())));
int errorOnlyCount = results.get(true).get(false).intValue();
int warningOnlyCount = results.get(false).get(true).intValue();
int bothErrorAndWarningCount = results.get(true).get(true).intValue();
int successfullyProcessedCount = results.get(false).get(false).intValue();

From the particular structure of this code, it looks like you could eliminate the map conversion entirely and partition directly on getIsDuplicateItemError and getIsDescriptionWarning, but this is most similar to the code you've already written.