Injection point not found

[pgtaboada]

Hi,

I have a CSV file with this header:

Patientennummer;Matchcode;Firma;Anrede;Titel;Name;Anschrift;Land;Briefanrede;Behandelt;Telefon1;Telefon2;Telefax;Mobil;Email1;Email2;Geschlecht;Geboren;Datum;Geändert;R_Geboren;Kontoinhaber;Bank;Konto;BLZ;IBAN;BIC;Internet;PatNr;R_Firma1;R_Anrede;R_Titel;R_Name;R_Anschrift;R_Land;R_Briefanrede;Beruf;Familienstand;Größe;Gewicht;Bemerkung;Diagnose;Anamnese;Hinweis;Versicherung;Risikofaktoren;Allergien;Unfälle;Operationen;Dauertherapie;Versicherer;Zahlungsziel;

And I am trying to parse using:

CsvParser .separator(';') .mapTo(SmedPatient.class) .defaultHeaders() .stream(isr) .forEach(p -> { repo.getPatienten().put(p.parsedPatientNummer(), p); });

I have the property annotated as following:

@Column(name="Geändert") private String geaendert;

What am I doing wrong? Is it the umlauts?

[arnaudroger]

The umlauts should not be an issue, it's should all be UTF-8. What's the @Column package? I assume there a setter for geaendert ?

[arnaudroger]

Quick question, why did you use .defaultHeaders(), it probably not do what you think it those, I should deprecate it ... try without it

[arnaudroger]

so the @Column if that the one from javax.persistence wont work without sfm-jdbc - it provides the AliasProvider hooks for it-. There is another way to specify an alias. it's not as straightforward as it should be - will fix that -.

If you use that

CsvParser.separator(';')
    .mapWith(
       CsvMapperFactory
            .newInstance()
            .addAlias("Geändert", "geaendert" )
            .newMapper(SmedPatient.class))
    .stream(isr) .forEach(p -> { repo.getPatienten().put(p.parsedPatientNummer(), p); });

I think it should work

[arnaudroger]

the next release will be easier

CsvParser.separator(';')
    .mapTo(C598.class)
    .alias("Geändert", "geaendert")
    .stream(isr) .forEach(p -> { repo.getPatienten().put(p.parsedPatientNummer(), p); });

[pgtaboada]

Unfortunately, the File is not UTF-8, but I am reading it accordingly:

new InputStreamReader(stream, StandardCharsets.ISO_8859_1)

The Java-Files and therefore the annotation name is indeed UTF-8.

I Removed the "defaultHeaders()" from the configuration, different Problem:

Injection point for on class de.(...).Patient not found

I did not remove the content of "for on class".

[arnaudroger]

have you tried with the mapWith, I created a test for that and with it working?

[arnaudroger]

though injection point is not the error I got. Which version are you running?

[pgtaboada]

I had a very old Version, switched to compile 'org.simpleflatmapper:sfm-csv:6.0.11' but it did not really solve the problem:

org.simpleflatmapper.map.MapperBuildingException: Could not find eligible property for 'Geändert' on class de.(...)SmedPatient not found See https://github.com/arnaudroger/SimpleFlatMapper/wiki/Errors_PROPERTY_NOT_FOUND

I did not try mapWith, I have 52 properties... :)

[arnaudroger]

is the headers part of the file? then you won't need to specify the property, just need to add the alias. or are you saying you have more than one @Column ?

using mapWith does not mean you are providing a static mapper. the CsvMapperFactory will return a dynamic one which will use the first row as headers.

[arnaudroger]

also you try adding sfm-jdbc to the dependency, that should add support for the javax.persistence.Column

[pgtaboada]

The headers are part of the file, I'll try this way

[pgtaboada]

I found a Problem - the header in the file ends with an ";"...

The CsvMapper thinks this is a new "" column and fails.

Unfortunately, the file in question comes from a third party tool and I cannot fix the header.

[arnaudroger]

Hum it should prob handle that. In the mean time there is a work around I’ll get the code You’ll need to

addColumnProperty(« « , ignoreproperty.instance) or similar on the mapper factory

[arnaudroger]

with the CsvMapperFactory

CsvParser.separator(';')
    .mapWith(
       CsvMapperFactory
            .newInstance()
            .addAlias("Geändert", "geaendert" )
        .addColumnProperty("", new IgnoreProperty())  
            .newMapper(SmedPatient.class))
    .stream(isr) .forEach(p -> { repo.getPatienten().put(p.parsedPatientNummer(), p); });

if the @Column works

CsvParser
    .separator(';') 
    .mapTo(SmedPatient.class)
    .columnDefinition("", CsvColumnDefinition.ignoreDefinition())
    .stream(isr) .forEach(p -> { repo.getPatienten().put(p.parsedPatientNummer(), p); });

I'll fix that as well in the next release that should work out of the box

[arnaudroger]

just pushed 6.0.13 to maven central. that should reduce the code needed to just

CsvParser.separator(';')
    .mapTo(C598.class)
    .alias("Geändert", "geaendert")
    .stream(isr) .forEach(p -> { repo.getPatienten().put(p.parsedPatientNummer(), p); });

has it now handles empty trailing headers

[pgtaboada]

Awesome. It worked perfectly after the tips you gave, this is a great lib.

Can you tell how long it takes until it is visible in mvn?

[arnaudroger]

it's there now. thanks for the ticket, those kinds of feedback helps a lot ironing out issues I did not encounter or thought about.