1
0
Fork 0
mirror of https://github.com/documize/community.git synced 2025-07-19 05:09:42 +02:00
documize/domain/conversion/conversion.go

339 lines
9.1 KiB
Go
Raw Normal View History

2017-08-02 12:39:12 +01:00
// Copyright 2016 Documize Inc. <legal@documize.com>. All rights reserved.
//
// This software (Documize Community Edition) is licensed under
// GNU AGPL v3 http://www.gnu.org/licenses/agpl-3.0.en.html
//
// You can operate outside the AGPL restrictions by purchasing
// Documize Enterprise Edition and obtaining a commercial license
// by contacting <sales@documize.com>.
//
// https://documize.com
package conversion
import (
"bytes"
"encoding/hex"
"fmt"
"io"
"net/http"
"path/filepath"
2017-08-02 12:39:12 +01:00
"strings"
api "github.com/documize/community/core/convapi"
"github.com/documize/community/core/env"
2017-08-02 12:39:12 +01:00
"github.com/documize/community/core/request"
"github.com/documize/community/core/response"
"github.com/documize/community/core/stringutil"
"github.com/documize/community/core/uniqueid"
"github.com/documize/community/domain"
ls "github.com/documize/community/domain/conversion/store"
2017-09-18 17:53:42 +01:00
"github.com/documize/community/domain/permission"
indexer "github.com/documize/community/domain/search"
"github.com/documize/community/domain/store"
2017-08-02 12:39:12 +01:00
"github.com/documize/community/model/activity"
"github.com/documize/community/model/attachment"
"github.com/documize/community/model/audit"
cm "github.com/documize/community/model/category"
2017-08-02 12:39:12 +01:00
"github.com/documize/community/model/doc"
"github.com/documize/community/model/page"
"github.com/documize/community/model/space"
"github.com/documize/community/model/workflow"
2017-08-02 12:39:12 +01:00
uuid "github.com/nu7hatch/gouuid"
"github.com/pkg/errors"
)
2017-08-02 15:26:31 +01:00
var storageProvider StorageProvider
2017-08-02 12:39:12 +01:00
func init() {
storageProvider = new(ls.LocalStorageProvider)
2017-08-02 12:39:12 +01:00
}
func (h *Handler) upload(w http.ResponseWriter, r *http.Request) (string, string, string) {
method := "conversion.upload"
ctx := domain.GetRequestContext(r)
spaceID := request.Param(r, "spaceID")
2017-08-02 12:39:12 +01:00
if !permission.CanUploadDocument(ctx, *h.Store, spaceID) {
2017-08-02 12:39:12 +01:00
response.WriteForbiddenError(w)
return "", "", ""
}
// grab file
filedata, filename, err := r.FormFile("attachment")
if err != nil {
response.WriteMissingDataError(w, method, "attachment")
return "", "", ""
}
b := new(bytes.Buffer)
_, err = io.Copy(b, filedata)
if err != nil {
response.WriteServerError(w, method, err)
2017-08-03 10:00:24 +01:00
h.Runtime.Log.Error(method, err)
2017-08-02 12:39:12 +01:00
return "", "", ""
}
// generate job id
newUUID, err := uuid.NewV4()
if err != nil {
response.WriteServerError(w, method, err)
2017-08-03 10:00:24 +01:00
h.Runtime.Log.Error(method, err)
2017-08-02 12:39:12 +01:00
return "", "", ""
}
job := newUUID.String()
err = storageProvider.Upload(job, filename.Filename, b.Bytes())
if err != nil {
response.WriteServerError(w, method, err)
2017-08-03 10:00:24 +01:00
h.Runtime.Log.Error(method, err)
2017-08-02 12:39:12 +01:00
return "", "", ""
}
h.Runtime.Log.Info(fmt.Sprintf("Org %s (%s) [Uploaded] %s", ctx.OrgName, ctx.OrgID, filename.Filename))
return job, spaceID, ctx.OrgID
2017-08-02 12:39:12 +01:00
}
func (h *Handler) convert(w http.ResponseWriter, r *http.Request, job, spaceID string, conversion api.ConversionJobRequest) {
2017-08-02 12:39:12 +01:00
method := "conversion.upload"
ctx := domain.GetRequestContext(r)
2017-08-27 16:39:09 +01:00
licenseKey, _ := h.Store.Setting.Get("EDITION-LICENSE", "key")
licenseSignature, _ := h.Store.Setting.Get("EDITION-LICENSE", "signature")
2017-08-02 12:39:12 +01:00
k, _ := hex.DecodeString(licenseKey)
s, _ := hex.DecodeString(licenseSignature)
conversion.LicenseKey = k
conversion.LicenseSignature = s
org, err := h.Store.Organization.GetOrganization(ctx, ctx.OrgID)
if err != nil {
response.WriteServerError(w, method, err)
2017-08-03 10:00:24 +01:00
h.Runtime.Log.Error(method, err)
2017-08-02 12:39:12 +01:00
return
}
conversion.ServiceEndpoint = org.ConversionEndpoint
var fileResult *api.DocumentConversionResponse
var filename string
filename, fileResult, err = storageProvider.Convert(conversion)
if err != nil {
response.WriteServerError(w, method, err)
2017-08-03 10:00:24 +01:00
h.Runtime.Log.Error(method, err)
2017-08-02 12:39:12 +01:00
return
}
if fileResult.Err != "" {
response.WriteServerError(w, method, errors.New(fileResult.Err))
2017-08-03 10:00:24 +01:00
h.Runtime.Log.Error(method, err)
2017-08-02 12:39:12 +01:00
return
}
// NOTE: empty .docx documents trigger this error
if len(fileResult.Pages) == 0 {
response.WriteMissingDataError(w, method, "no pages in document")
return
}
ctx.Transaction, err = h.Runtime.Db.Beginx()
if err != nil {
response.WriteServerError(w, method, err)
2017-08-03 10:00:24 +01:00
h.Runtime.Log.Error(method, err)
2017-08-02 12:39:12 +01:00
return
}
// Fetch space where document resides.
sp, err := h.Store.Space.Get(ctx, spaceID)
if err != nil {
ctx.Transaction.Rollback()
response.WriteServerError(w, method, err)
h.Runtime.Log.Error(method, err)
return
}
nd, err := processDocument(ctx, h.Runtime, h.Store, h.Indexer, filename, job, sp, fileResult)
2017-08-02 12:39:12 +01:00
if err != nil {
ctx.Transaction.Rollback()
2017-08-02 12:39:12 +01:00
response.WriteServerError(w, method, err)
2017-08-03 10:00:24 +01:00
h.Runtime.Log.Error(method, err)
2017-08-02 12:39:12 +01:00
return
}
response.WriteJSON(w, nd)
}
func processDocument(ctx domain.RequestContext, r *env.Runtime, store *store.Store, indexer indexer.Indexer, filename,
job string, sp space.Space, fileResult *api.DocumentConversionResponse) (newDocument doc.Document, err error) {
2017-08-02 12:39:12 +01:00
// Convert into database objects
document := convertFileResult(filename, fileResult)
document.Job = job
document.OrgID = ctx.OrgID
2018-09-19 16:03:29 +01:00
document.SpaceID = sp.RefID
2017-08-02 12:39:12 +01:00
document.UserID = ctx.UserID
documentID := uniqueid.Generate()
document.RefID = documentID
if r.Product.Edition == domain.CommunityEdition {
document.Lifecycle = workflow.LifecycleLive
} else {
document.Lifecycle = sp.Lifecycle
}
2017-08-02 12:39:12 +01:00
err = store.Document.Add(ctx, document)
if err != nil {
ctx.Transaction.Rollback()
err = errors.Wrap(err, "cannot insert new document")
return
}
for k, v := range fileResult.Pages {
var p page.Page
p.OrgID = ctx.OrgID
p.DocumentID = documentID
p.Level = v.Level
2018-09-19 16:03:29 +01:00
p.Name = v.Title
2017-08-02 12:39:12 +01:00
p.Body = string(v.Body)
p.Sequence = float64(k+1) * 1024.0 // need to start above 0 to allow insertion before the first item
pageID := uniqueid.Generate()
p.RefID = pageID
p.ContentType = "wysiwyg"
2018-09-19 16:03:29 +01:00
p.Type = "section"
2017-08-02 12:39:12 +01:00
meta := page.Meta{}
2018-09-19 16:03:29 +01:00
meta.SectionID = pageID
2017-08-02 12:39:12 +01:00
meta.RawBody = p.Body
meta.Config = "{}"
model := page.NewPage{}
model.Page = p
model.Meta = meta
err = store.Page.Add(ctx, model)
if err != nil {
ctx.Transaction.Rollback()
err = errors.Wrap(err, "cannot insert new page for new document")
return
}
// pp, _ := store.Page.Get(ctx, pageID)
go indexer.IndexContent(ctx, p)
2017-08-02 12:39:12 +01:00
}
da := []attachment.Attachment{}
2017-08-02 12:39:12 +01:00
for _, e := range fileResult.EmbeddedFiles {
//fmt.Println("DEBUG embedded file info", document.OrgId, document.Job, e.Name, len(e.Data), e.ID)
var a attachment.Attachment
a.DocumentID = documentID
a.Job = document.Job
a.FileID = e.ID
a.Filename = strings.Replace(e.Name, "embeddings/", "", 1)
a.Data = e.Data
refID := uniqueid.Generate()
a.RefID = refID
err = store.Attachment.Add(ctx, a)
if err != nil {
ctx.Transaction.Rollback()
err = errors.Wrap(err, "cannot insert attachment for new document")
return
}
da = append(da, a)
2017-08-02 12:39:12 +01:00
}
// Add default categories to newly created document (if we have them).
cats, err := store.Category.GetBySpace(ctx, document.SpaceID)
if err != nil {
r.Log.Error("fetch default categories for new document", err)
}
for ic := range cats {
if cats[ic].IsDefault {
c := cm.Member{}
c.OrgID = ctx.OrgID
c.SpaceID = sp.RefID
c.RefID = uniqueid.Generate()
c.DocumentID = document.RefID
c.CategoryID = cats[ic].RefID
err = store.Category.AssociateDocument(ctx, c)
if err != nil {
r.Log.Error("apply default category to new document", err)
}
}
}
2017-08-21 07:43:18 +01:00
store.Activity.RecordUserActivity(ctx, activity.UserActivity{
SpaceID: document.SpaceID,
DocumentID: document.RefID,
2017-08-21 07:43:18 +01:00
SourceType: activity.SourceTypeDocument,
ActivityType: activity.TypeCreated})
err = ctx.Transaction.Commit()
if err != nil {
err = errors.Wrap(err, "cannot commit new document import")
return
}
2017-08-18 14:31:36 +01:00
2017-08-02 12:39:12 +01:00
newDocument, err = store.Document.Get(ctx, documentID)
if err != nil {
err = errors.Wrap(err, "cannot fetch new document")
return
}
2018-10-10 15:13:09 +01:00
go indexer.IndexDocument(ctx, newDocument, da)
store.Space.SetStats(ctx, newDocument.SpaceID)
2017-08-02 12:39:12 +01:00
store.Audit.Record(ctx, audit.EventTypeDocumentUpload)
return
}
// convertFileResult takes the results of a document upload and convert,
// and creates the outline of a database record suitable for inserting into the document
// table.
func convertFileResult(filename string, fileResult *api.DocumentConversionResponse) (document doc.Document) {
document = doc.Document{}
document.RefID = ""
document.OrgID = ""
2018-09-19 16:03:29 +01:00
document.SpaceID = ""
2017-08-02 12:39:12 +01:00
document.Job = ""
document.Location = filename
// Make document name from filename minus extension.
document.Name = GetDocumentNameFromFilename(filename)
document.Slug = stringutil.MakeSlug(document.Name)
2017-08-02 12:39:12 +01:00
if fileResult != nil {
document.Excerpt = fileResult.Excerpt
}
document.Tags = "" // now a # separated list of tag-words, rather than JSON
return document
}
// GetDocumentNameFromFilename strips path and extension.
func GetDocumentNameFromFilename(filename string) (dn string) {
dn = filename
// First try Linux separator.
t := strings.SplitAfter(filename, "/")
if len(t) > 1 {
dn = t[len(t)-1]
} else {
// Now try Linux separator.
t = strings.SplitAfter(filename, "\\")
if len(t) > 1 {
dn = t[len(t)-1]
}
}
// Remove file extension.
dn = strings.TrimSuffix(dn, filepath.Ext(dn))
return
}